US20130318269A1 - Processing structured and unstructured data using offload processors - Google Patents

Processing structured and unstructured data using offload processors Download PDF

Info

Publication number
US20130318269A1
US20130318269A1 US13900333 US201313900333A US2013318269A1 US 20130318269 A1 US20130318269 A1 US 20130318269A1 US 13900333 US13900333 US 13900333 US 201313900333 A US201313900333 A US 201313900333A US 2013318269 A1 US2013318269 A1 US 2013318269A1
Authority
US
Grant status
Application
Patent type
Prior art keywords
memory
data
ximm
processing
modules
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Abandoned
Application number
US13900333
Inventor
Parin Bhadrik Dalal
Stephen Paul Belair
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
XOCKETS Inc
Xockets IP LLC
Original Assignee
Xockets IP LLC
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date

Links

Images

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • H04L47/24Flow control or congestion control depending on the type of traffic, e.g. priority or quality of service [QoS]
    • H04L47/2441Flow classification
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/16Handling requests for interconnection or transfer for access to memory bus
    • G06F13/1605Handling requests for interconnection or transfer for access to memory bus based on arbitration
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/16Handling requests for interconnection or transfer for access to memory bus
    • G06F13/1605Handling requests for interconnection or transfer for access to memory bus based on arbitration
    • G06F13/1652Handling requests for interconnection or transfer for access to memory bus based on arbitration in a multiprocessor architecture
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/16Handling requests for interconnection or transfer for access to memory bus
    • G06F13/1668Details of memory controller
    • G06F13/1673Details of memory controller using buffers
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/20Handling requests for interconnection or transfer for access to input/output bus
    • G06F13/28Handling requests for interconnection or transfer for access to input/output bus using burst mode transfer, e.g. direct memory access DMA, cycle steal
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/14Handling requests for interconnection or transfer
    • G06F13/36Handling requests for interconnection or transfer for access to common bus or bus system
    • G06F13/362Handling requests for interconnection or transfer for access to common bus or bus system with centralised access control
    • G06F13/364Handling requests for interconnection or transfer for access to common bus or bus system with centralised access control using independent requests or grants, e.g. using separated request and grant lines
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/38Information transfer, e.g. on bus
    • G06F13/40Bus structure
    • G06F13/4004Coupling between buses
    • G06F13/4022Coupling between buses using switching circuits, e.g. switching matrix, connection or expansion network
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/38Information transfer, e.g. on bus
    • G06F13/40Bus structure
    • G06F13/4004Coupling between buses
    • G06F13/4027Coupling between buses using bus bridges
    • G06F13/404Coupling between buses using bus bridges with address mapping
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F13/00Interconnection of, or transfer of information or other signals between, memories, input/output devices or central processing units
    • G06F13/38Information transfer, e.g. on bus
    • G06F13/42Bus transfer protocol, e.g. handshake; Synchronisation
    • G06F13/4204Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus
    • G06F13/4234Bus transfer protocol, e.g. handshake; Synchronisation on a parallel bus being a memory bus
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F21/00Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
    • G06F21/50Monitoring users, programs or devices to maintain the integrity of platforms, e.g. of processors, firmware or operating systems
    • G06F21/55Detecting local intrusion or implementing counter-measures
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L29/00Arrangements, apparatus, circuits or systems, not covered by a single one of groups H04L1/00 - H04L27/00 contains provisionally no documents
    • H04L29/02Communication control; Communication processing contains provisionally no documents
    • H04L29/06Communication control; Communication processing contains provisionally no documents characterised by a protocol
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic regulation in packet switching networks
    • H04L47/10Flow control or congestion control
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L49/00Packet switching elements
    • H04L49/70Virtual switches
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L63/00Network architectures or network communication protocols for network security
    • H04L63/02Network architectures or network communication protocols for network security for separating internal from external traffic, e.g. firewalls
    • H04L63/0227Filtering policies
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing
    • Y02D10/10Reducing energy consumption at the single machine level, e.g. processors, personal computers, peripherals or power supply
    • Y02D10/14Interconnection, or transfer of information or other signals between, memories, peripherals or central processing units
    • YGENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02DCLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THIR OWN ENERGY USE
    • Y02D10/00Energy efficient computing
    • Y02D10/10Reducing energy consumption at the single machine level, e.g. processors, personal computers, peripherals or power supply
    • Y02D10/15Reducing energy consumption at the single machine level, e.g. processors, personal computers, peripherals or power supply acting upon peripherals
    • Y02D10/151Reducing energy consumption at the single machine level, e.g. processors, personal computers, peripherals or power supply acting upon peripherals the peripheral being a bus

Abstract

Methods of processing structured data are disclosed that can include providing a plurality of XIMM modules connected to a memory bus in a first server, with the XIMM modules each respectively having a DMA slave module connected to the memory bus and an arbiter for scheduling tasks, with the XIMM modules further providing an in-memory database; and connecting a central processing unit (CPU) in the first server to the XIMM modules by the memory bus, with the CPU arranged to process and direct structured queries to the plurality of XIMM modules.

Description

    PRIORITY CLAIMS
  • This application claims the benefit of U.S. Provisional Patent Applications 61/650,373 filed May 22, 2012, 61/753,892 filed on Jan. 17, 2013, 61/753,895 filed on Jan. 17, 2013, 61/753,899 filed on Jan. 17, 2013, 61/753,901 filed on Jan. 17, 2013, 61/753,903 filed on Jan. 17, 2013, 61/753,904 filed on Jan. 17, 2013, 61/753,906 filed on Jan. 17, 2013, 61/753,907 filed on Jan. 17, 2013, and 61/753,910 filed on Jan. 17, 2013, the contents all of which are incorporated by reference herein.
  • TECHNICAL FIELD
  • The present disclosure relates generally to servers capable of efficiently processing structured and unstructured data. More particularly, systems supporting offload or auxiliary processing modules that can be physically connected to a system memory bus to process data independent of a host processor of the server are described.
  • BACKGROUND
  • Enterprises store and process their large amounts of data in a variety of ways. One manner in which enterprises store data is by using relational databases and corresponding relational database management systems (RDBMS). Such data, usually referred to as structured data, may be collected, normalized, formatted and stored in an RDBMS. Tools based on standardized data languages such as the Structured Query Language (SQL) may be used for accessing and processing structured data. However, it is estimated that such formatted structured data represents only a tiny fraction of an enterprise's stored data. Organizations are becoming increasingly aware that substantial information and knowledge resides in unstructured data (i.e., “Big Data”) repositories. Accordingly, simple and effective access to both structured and unstructured data are seen as necessary for maximizing the value of enterprise informational resources.
  • However, conventional platforms that are currently being used to handle structured and unstructured data can substantially differ in their architecture. In-memory processing and Storage Area Network (SAN)-like architectures are used for traditional SQL queries, while commodity or shared nothing architectures (each computing node, consisting of a processor, local memory, and disk resources, shares nothing with other nodes in the computing cluster) are usually used for processing unstructured data. An architecture that supports both structured and unstructured queries can better handle current and emerging Big Data applications.
  • SUMMARY
  • A method of processing structured data include providing a plurality of XIMM modules connected to a memory bus in a first server, with the XIMM modules each respectively having a DMA slave module connected to the memory bus and an arbiter for scheduling tasks, with the XIMM modules further providing an in-memory database; and connecting a central processing unit (CPU) in the first server to the XIMM modules by the memory bus, with the CPU arranged to process and direct structured queries to the plurality of XIMM modules.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • FIG. 1 shows illustrates an embodiment suitable to process structured queries.
  • FIGS. 2-1 and 2-2 are diagrams showing the workflow and distributed architecture used to implement a data processing software.
  • FIG. 3-1 is flow diagram showing a data processing method according to an embodiment.
  • FIG. 3-2 shows a data processing architecture according to an embodiment.
  • FIG. 4-1 shows a cartoon schematically illustrating a data processing system according to an embodiment, including a removable computation module for offload of data processing.
  • FIG. 4-2 shows an example layout of an in-line module (referred to as a “XIMM”) module according to an embodiment.
  • FIG. 4-3 shows two possible architectures for a XIMM in a simulation (Xockets MAX and MIN).
  • FIG. 4-4 shows a representative the power budget for a XIMMs according to various embodiments.
  • FIG. 4-5 illustrates data flow operation of one embodiment using an ARM A9 architecture.
  • FIG. 5-1 is a block schematic diagram of a processing module according to an embodiment.
  • FIGS. 5-2 and 5-3 are diagrams of a processor module according to embodiments.
  • FIG. 5-4 is a diagram showing an opposing side of a processor module like that of FIG. 5-2 or 5-3, according to an embodiment.
  • FIG. 5-5 is a diagram of a system according to an embodiment.
  • FIGS. 5-6 to 5-11 are block schematic diagrams showing processor module operations according to particular embodiments.
  • FIG. 5-12 is a flow diagram of a method according to an embodiment.
  • FIG. 5-13 is a flow diagram of a method according to another embodiment.
  • FIG. 6-1 is a block schematic diagram of a system according to another embodiment.
  • FIG. 6-2 is a diagram showing a system flow according to an embodiment.
  • DETAILED DESCRIPTION
  • Data processing and analytics for enterprise server or cloud based data systems, including both structured or unstructured data, can be efficiently implemented on offload processing modules connected to a memory bus, for example, by insertion into a socket for a Dual In-line Memory Module (DIMM). Such modules can be referred to as Xocket™ In-line Memory Modules (XIMMs), and can have multiple “wimpy” cores associated with a memory channel. Using one or more XIMMs it is possible to execute lightweight data processing tasks without intervention from a main server processor. As will be discussed, XIMM modules have high efficiency context switching, high parallelism, and can efficiently process large data sets. Such systems as a whole are able to handle large database searches at a very low power when compared to traditional high power “brawny” server cores. Advantageously, by accelerating implementation of MapReduce or similar algorithms on unstructured data and by providing high performance virtual shared disk for structured queries, a XIMM based architecture capable of partitioning tasks is able to greatly improve data analytic performance.
  • FIG. 1 illustrates an embodiment to process structured queries according to an embodiment. FIG. 1 depicts two commodity computers which can be rack servers (140,140 a), each of which includes at least one central processing unit (108) and a set of offload processors (112 a to 112 c). The servers (140, 140 a) are preferably present in the same rack and connected by a top of rack (TOR) switch (120). The servers (140, 140 a) are suited for hosting a distributed and shared in-memory assembly (110 a, 110 b) that can handle and respond to structured queries as one large in-memory system.
  • Traditionally, an in-memory database that handles structured queries looks for the query in its physical memory, and in the absence of the query therein will perform a disk read to a database. The disk read might result in the page cache of the kernel being populated with the disk file. The process might perform a complete copy of the page(s) (in a paged memory system) into its process buffer or it might perform a mmap operation, wherein a pointer to the entry in the page cache storing the pages is created and stored in the heap of the process. The latter is less time consuming and more efficient than the former.
  • The inclusion of a XIMM supported data retrieval framework can effectively extend the overall available size of the in-memory space available with the assembly. In case a structured query is being handled by one of the said CPUs (108), and the CPU is unable to retrieve it from its main memory, it will immediately trap into a memory map (mmap routine) that will handle disk reads and populate the page cache.
  • The embodiment described herein can modify the mmap routine of standard operating systems to execute code corresponding to a driver for a removable computation module driver, in this case a XIMM driver. The XIMM driver in turn identifies the query and transfers the query to one or more XIMMs in the form of memory reads/writes. A XIMM can house a plurality of offload processors (112 a to 112 c) that can receive the memory read/write commands containing the structured query. One or more of the offload processors (112 a to 112 c) can perform a search for the query in its available cache and local memory and return the result. According to embodiments, an mmap query can further be modified to allow the transference of the query to XIMMs that are not in the same server but in the same rack. The mmap abstraction in such a case can perform a remote direct memory access (RDMA) or a similar network memory read for accessing data present in a XIMM that is in the same rack. As a second XIMM (i.e., offload processors 112 a to 112 c of server 140 a) is connected to a first XIMM (i.e., offload processors 112 a to 112 c of server 140) through a top of rack switch 120, the latency of a system according to an embodiment can be the combined latency of the network interface cards (NICs) of the two servers (i.e., 100 in 140 and 100 in 140 a), the latency of the TOR switch 120, and a response time of the second XIMM.
  • Despite the additional hops, embodiments can provide a system that can have a much lower latency than a read from a hard disk drive. The described architecture can provide orders of magnitude improvement over a single structured query in-memory database. By allowing for non-frequently used data of one of the servers to be pushed to another XIMM located in the same rack, the effective in-memory space can be increased beyond conventional limits, and a large assembly having a large physical, low latency memory space can be created. This embodiment may further be improved by allowing for transparent sharing of pages across multiple XIMMs. Embodiments can further improve the latency between server-to-server connections by mediating them by XIMMs. The XIMMs can act as intelligent switches to offload and bypass the TOR switch.
  • Conventional data intensive computing platforms for handling large volumes of unstructured data can use a parallel computing approach combining multiple processors and disks in large commodity computing clusters connected with high-speed communications switches and networks. This can allow the data to be partitioned among the available computing resources and processed independently to achieve performance and scalability based on the amount of data. A variety of distributed architectures have been developed for data-intensive computing and several software frameworks have been proposed to process unstructured data. One such programming model for processing large data sets with a parallel, distributed algorithm on a multiple servers or clusters is commonly known as MapReduce. Hadoop is a popular open-source implementation of MapReduce that is widely used by enterprises for search of unstructured data.
  • FIGS. 2-1 and 2-2 show the workflow and distributed architecture used to implement a data processing software, such as Hadoop. Referring to FIG. 2-1, Hadoop MapReduce workloads may be broken into Split, Map, Shuffle, Reduce and Merge phases. The input file (204) is fetched from a file system such as Hadoop Distributed File System (HDFS) (202) and divided into smaller pieces referred to as Splits. Each Split (206) is a contiguous range of the input. Each record in each input split is independently passed to a Map function run by a Mapper (208) hosted on a processor. The Map function accepts a single record as an input and produces zero or more output records, each of which contains a key and a value. In the Shuffle phase (210), the results from the Map functions, referred to as intermediate (key, value) pairs, are rearranged such that all values with the same key are grouped together. The Reduce function run by a Reducer (212) takes in a key value and a list values as the input and produces another list as the output. The output records from Reduce functions are merged to form an Output file (216) which is then stored in the file system (202).
  • Map and Reduce tasks are computationally intensive and have very tight dependency on each other. While Map tasks are small and independent tasks that can run in parallel, the Reduce tasks include fetching intermediate (key, value) pairs that result from each Map function, sorting and merging intermediate results according to keys and applying Reduce functions to the sorted intermediate results. Reducer (212) can perform the Reduce function only after it receives the intermediate results from all the Mappers (208). Thus, the Shuffle step (210) (communicating the Map results to Reducers) often becomes the bottleneck in Hadoop workloads and introduces latency.
  • FIG. 2-2 shows a typical computing cluster used to implement Hadoop. A Master Node (222) runs a JobTracker (224) which organizes the cluster's activities. Each of the Worker Nodes (326) run a TaskTracker (228) which organizes the worker node's activities. All input jobs are organized into sequential tiers of tasks (230). The tasks could be map tasks or reduce tasks. The TaskTracker (228) runs a number of map and reduce tasks concurrently and pulls new tasks from the JobTracker (224) as soon as the old tasks are completed. The Hadoop Map-Reduce layer stores intermediate data (232) produced by the map and reduce tasks in the Hadoop Distributed File System (202) discussed in FIG. 2-1. HDFS (202) is designed to provide high streaming throughput to large, write-once-read-many-times files.
  • In an embodiment, a XIMM based architecture can improve Hadoop (or similar data processing) performance in two ways. Firstly, intrinsically parallel computational tasks can be allocated to the XIMMs (e.g., modules with offload processors), leaving the number crunching tasks to “brawny” (e.g., x86) cores. This is illustrated in more detail in FIG. 3-1.
  • FIG. 3-1 is flow diagram showing a data processing method according to an embodiment. A method can start (302) and input data can be fetched from a file system (304). In particular embodiments this can include a Direct Memory Access (DMA) operation from an HDFS. In some embodiments, all DMA operations are engineered such that all the parallel Map steps are equitably served with data. The input data can be partitioned into splits in (306) and parsed into records that contain initial (key, value) pairs in (308). Since the tasks involved in steps (304, 306 and 308) are computationally light, they can be performed all, or in part, by the offload processors hosted on the XIMMs (e.g., by wimpy cores). Map operations can then be performed on the initial (key, value) pairs (310). The intermediate (key, value) pairs that result from the Map operations can then be communicated to the Reducers (312). Once the results from all the Map operations are available, Reduce operations can be performed (314). The results from the Reducers are merged into a single output file and written back to the file system (e.g., 302) by offload processors (316). Since Map and Reduce functions are computationally intensive, the steps (310 and 314) can be handled by a CPU (e.g., by brawny cores). Such distribution of workloads to processor cores that are favorably disposed to perform them can reduce latency and/or increases efficiency.
  • A XIMM based architecture, according to an embodiment, can reduce the intrinsic bottleneck of most Hadoop (or similar data processing) workloads (the Shuffle phase) by driving the I/O backplane to its full capacity. In a conventional Hadoop system, the TaskTracker (228) described in FIG. 2-2 serves hyper-text transport protocol (HTTP) GET requests to communicate Map results to Reduce inputs. Also, the Mappers and Reducers hosted by the individual server CPUs have to communicate using their corresponding top of rack (TOR) switches of the corresponding racks. This slows down the process. Further, the Reducers hosted by the CPU stay idle while the results from Map operations are collected and sorted, introducing further computational inefficiency.
  • FIG. 3-2 shows a data processing architecture according to an embodiment. In the architecture illustrated in FIG. 3-2, operational modules are labeled to reflect their primary operations as previously noted with respect to FIG. 3-1. In an embodiment, instead of using HTTP to communicate, a publish-subscribe model is used to perform the Shuffling phase as shown in FIG. 3-2. Since the results from the Map step (310) are already stored and available in main memory, they can be collected through DMA operations (312 a) and parsed (312 b) by XIMMs. Since Map results are not required to be written to disk, latency can be reduced. Further, the XIMMs incorporated into the individual servers can define a switch fabric that can be used for Shuffling. The mid-plane defined by the XIMM based switch fabric is capable of driving and receiving the full 240 Gbps capacity of the PCI-3.0 bus and thus offers better speed and bandwidth compared to HTTP. The key is published through the massively parallel I/O mid-plane defined by the XIMMs (312 a). Subscriptions are identified based on the keys and are directed to the Reducers hosted on the CPU through virtual interrupts (312 b). Thus, rack-level locality and aggregation that are typical of conventional Hadoop systems are no longer required in the XIMM based architecture. Instead, intermediate (key, value) pairs are exchanged by all the computing nodes across several different servers through intelligent virtual switching of the XIMMs resulting in efficient processing of Hadoop workloads.
  • The following example(s) provide illustration and discussion of exemplary hardware and data processing systems suitable for implementation and operation of the foregoing discussed systems and methods. In particular hardware and operation of wimpy cores or computational elements connected to a memory bus and mounted in DIMM or other conventional memory socket is discussed.
  • FIG. 4-1 is a cartoon schematically illustrating a data processing system 400 including a removable computation module 402 for offload of data processing from x86 or similar main/server processors 403 to modules connected to a memory bus 403. Such modules 402 can be XIMM modules, as described herein or equivalents, and can have multiple computation elements that can be referred to as “offload processors” because they offload various “light touch” processing tasks such HTML, video, packet level services, security, or data analytics. This is of particular advantage for applications that require frequent random access or application context switching, since many server processors incur significant power usage or have data throughput limitations that can be greatly reduced by transfer of the computation to lower power and more memory efficient offload processors.
  • The computation elements or offload processors can be accessible through memory bus 405. In this embodiment, the module can be inserted into a Dual Inline Memory Module (DIMM) slot on a commodity computer or server using a DIMM connector (407), providing a significant increase in effective computing power to system 400. The module (e.g., XIMM) may communicate with other components in the commodity computer or server via one of a variety of busses including but not limited to any version of existing double data rate standards (e.g., DDR, DDR2, DDR3, etc.)
  • This illustrated embodiment of the module 402 contains five offload processors (400 a, 400 b, 400 c, 400 d, 400 e) however other embodiments containing greater or fewer numbers of processors are contemplated. The offload processors (400 a to 400 e) can be custom manufactured or one of a variety of commodity processors including but not limited to field-programmable grid arrays (FPGA), microprocessors, reduced instruction set computers (RISC), microcontrollers or ARM processors. The computation elements or offload processors can include combinations of computational FPGAs such as those based on Altera, Xilinx (e.g., Artix™ class or Zynq® architecture, e.g., Zynq® 7020), and/or conventional processors such as those based on Intel Atom or ARM architecture (e.g., ARM A9). For many applications, ARM processors having advanced memory handling features such as a snoop control unit (SCU) are preferred, since this can allow coherent read and write of memory. Other preferred advanced memory features can include processors that support an accelerator coherency port (ACP) that can allow for coherent supplementation of the cache through an FPGA fabric or computational element.
  • Each offload processor (400 a to 400 e) on the module 402 may run one of a variety of operating systems including but not limited to Apache or Linux. In addition, the offload processors (400 a to 400 e) may have access to a plurality of dedicated or shared storage methods. In this embodiment, each offload processor can connect to one or more storage units (in this embodiments, pairs of storage units 404 a, 404 b, 404 c and 404 d). Storage units (404 a to 404 d) can be of a variety of storage types, including but not limited to random access memory (RAM), dynamic random access memory (DRAM), sequential access memory (SAM), static random access memory (SRAM), synchronous dynamic random access memory (SDRAM), reduced latency dynamic random access memory (RLDRAM), flash memory, or other emerging memory standards such as those based on DDR4 or hybrid memory cubes (HMC).
  • FIG. 4-2 shows an example layout of a module (e.g., XIMM) such as that described in FIG. 4-1, as well as a connectivity diagram between the components of the module. In this example, five Xilinx™ Zynq® 7020 (416 a, 416 b, 416 c, 416 d, 416 e and 416 in the connectivity diagram) programmable systems-on-a-chip (SoC) are used as computational FPGAs/offload processors. These offload processors can communicate with each other using memory-mapped input-output (MMIO) (412). The types of storage units used in this example are SDRAM (SD, one shown as 408) and RLDRAM (RLD, three shown as 406 a, 406 b, 406 c) and an Inphi™ iMB02 memory buffer 418. Down conversion of 3.3 V to 2.5 volt is required to connect the RLDRAM (406 a to 406 c) with the Zynq® components. The components are connected to the offload processors and to each other via a DDR3 (414) memory bus. Advantageously, the indicated layout maximizes memory resources availability without requiring a violation of the number of pins available under the DIMM standard.
  • In this embodiment, one of the Zynq® computational FPGAs (416 a to 416 e) can act as arbiter providing a memory cache, giving an ability to have peer to peer sharing of data (via memcached or OMQ memory formalisms) between the other Zynq® computational FPGAs (416 a to 416 e). Traffic departing for the computational FPGAs can be controlled through memory mapped I/O. The arbiter queues session data for use, and when a computational FPGA asks for address outside of the provided session, the arbiter can be the first level of retrieval, external processing determination, and predictors set.
  • FIG. 4-3 shows two possible architectures for a module (e.g., XIMM) in a simulation (Xockets MAX and MIN). Xockets MIN (420 a) can be used in low-end public cloud servers, containing twenty ARM cores (420 b) spread across fourteen DIMM slots in a commodity server which has two Opteron x86 processors and two network interface cards (NICs) (420 c). This architecture can provide a minimal benefit per Watt of power used. Xockets MAX (422 a) contains eighty ARM cores (422 b) across eight DIMM slots, in a server with two Opteron x86 processors and four NICs (422 c). This architecture can provide a maximum benefit per Watt of power used.
  • FIG. 4-4 shows a representative power budget for an example of a module (e.g., XIMM) according to a particular embodiment. Each component is listed (424 a, 424 b, 424 c, 424 d) along with its power profile. Average total and total wattages are also listed (426 a, 426 b). In total, especially for I/O packet processing with packet sizes on the order 1 KB in size, module can have a low average power budget that is easily able to be provided by the 22 Vdd pins per DIMM. Additionally, the expected thermal output can be handled by inexpensive conductive heat spreaders, without requiring additional convective, conductive, or thermoelectric cooling. In certain situations, digital thermometers can be implemented to dynamically reduce performance (and consequent heat generation) if needed.
  • Operation of one embodiment of a module 430 (e.g., XIMM) using an ARM A9 architecture is illustrated with respect to FIG. 4-5. Use of ARM A9 architecture in conjunction with an FPGA fabric and memory, in this case shown as reduced latency DRAM (RLDRAM) 438, can simplify or makes possible zero-overhead context switching, memory compression and CPI, in part by allowing hardware context switching synchronized with network queuing. In this way, there can be a one-to-one mapping between thread and queues. As illustrated, the ARM A9 architecture includes a Snoop Control Unit 432 (SCU). This unit allows one to read out and write in memory coherently. Additionally, the Accelerator Coherency Port 434 (ACP) allows for coherent supplementation of the cache throughout the FPGA 436. The RLDRAM 438 provides the auxiliary bandwidth to read and write the ping-pong cache supplement (435): Block1$ and Block2$ during packet-level meta-data processing.
  • The following table (Table 1) illustrates potential states that can exist in the scheduling of queues/threads to XIMM processors and memory such as illustrated in FIG. 4-5.
  • TABLE 1
    Queue/Thread State HW treatment
    Waiting for Ingress All ingress data has been processed and thread
    Packet awaits further communication.
    Waiting for MMIO A functional call to MM hardware (such as HW
    encryption or transcoding) was made.
    Waiting for Rate-limit The thread's resource consumption exceeds limit,
    due to other connections idling.
    Currently being One of the ARM cores is already processing this
    processed thread, cannot schedule again.
    Ready for Selection The thread is ready for context selection.

    These states can help coordinate the complex synchronization between processes, network traffic, and memory-mapped hardware. When a queue is selected by a traffic manager a pipeline coordinates swapping in the desired L2 cache (440), transferring the reassembled 10 data into the memory space of the executing process. In certain cases, no packets are pending in the queue, but computation is still pending to service previous packets. Once this process makes a memory reference outside of the data swapped, a scheduler can require queued data from a network interface card (NIC) to continue scheduling the thread. To provide fair queuing to a process not having data, the maximum context size is assumed as data processed. In this way, a queue must be provisioned as the greater of computational resource and network bandwidth resource, for example, each as a ratio of an 800 MHz A9 and 3 Gbps of bandwidth. Given the lopsidedness of this ratio, the ARM core is generally indicated to be worthwhile for computation having many parallel sessions (such that the hardware's prefetching of session-specific data and TCP/reassembly offloads a large portion of the CPU load) and those requiring minimal general purpose processing of data.
  • Essentially zero-overhead context switching is also possible using modules as disclosed in FIG. 4-5. Because per packet processing has minimum state associated with it, and represents inherent engineered parallelism, minimal memory access is needed, aside from packet buffering. On the other hand, after packet reconstruction, the entire memory state of the session can be accessed, and so can require maximal memory utility. By using the time of packet-level processing to prefetch the next hardware scheduled application-level service context in two different processing passes, the memory can always be available for prefetching. Additionally, the FPGA 436 can hold a supplemental “ping-pong” cache (435) that is read and written with every context switch, while the other is in use. As previously noted, this is enabled in part by the SCU 432, which allows one to read out and write in memory coherently, and ACP 434 for coherent supplementation of the cache throughout the FPGA 436. The RLDRAM 438 provides for read and write to the ping-pong cache supplement (435): (shown as Block1$ and Block2$) during packet-level meta-data processing. In the embodiment shown, only locally terminating queues can prompt context switching.
  • In operation, metadata transport code can relieve a main or host processor from tasks including fragmentation and reassembly, and checksum and other metadata services (e.g., accounting, IPSec, SSL, Overlay, etc.). As IO data streams in and out, L1 cache 437 can be filled during packet processing. During a context switch, the lock-down portion of a translation lookaside buffer (TLB) of an L1 cache can be rewritten with the addresses corresponding to the new context. In one very particular implementation, the following four commands can be executed for the current memory space.
  • MRC p15,0,r0,c10,c0,0; read the lockdown register
  • BIC r0,r0,#1; clear preserve bit
  • MCR p15,0,r0,c10,c0,0; write to the lockdown register; write to the old value to the memory mapped Block RAM
  • This is a small 32 cycle overhead to bear. Other TLB entries can be used by the XIMM stochastically.
  • Bandwidths and capacities of the memories can be precisely allocated to support context switching as well as applications such as Openflow processing, billing, accounting, and header filtering programs.
  • For additional performance improvements, the ACP 434 can be used not just for cache supplementation, but hardware functionality supplementation, in part by exploitation of the memory space allocation. An operand can be written to memory and the new function called, through customizing specific Open Source libraries, so putting the thread to sleep and a hardware scheduler can validate it for scheduling again once the results are ready. For example, OpenVPN uses the OpenSSL library, where the encrypt/decrypt functions 439 can be memory mapped. Large blocks are then available to be exported without delay, or consuming the L2 cache 440, using the ACP 434. Hence, a minimum number of calls are needed within the processing window of a context switch, improving overall performance.
  • FIG. 5-1 is a block diagram of a processing module 500 according to another embodiment. A processing module 500 can be one implementation of XIMM as described herein. A processing module 500 can include a physical in-line module connector 502, a memory interface 504, arbiter logic 506, offload processor(s) 508, local memory 510, and control logic 512. A connector 502 can provide a physical connection to system memory bus. This is in contrast to a host processor which can access a system memory bus via a memory controller, or the like. In very particular embodiments, a connector 502 can be compatible with a dual in-line memory module (DIMM) slot of a computing system. Accordingly, a system including multiple DIMM slots can be populated with one or more processing modules 500, or a mix of processing modules and DIMM modules.
  • A memory interface 504 can detect data transfers on a system memory bus, and in appropriate cases, enable write data to be stored in the processing module 500 and/or read data to be read out from the processing module 500. In some embodiments, a memory interface 504 can be a slave interface, thus data transfers are controlled by a master device separate from the processing module. In very particular embodiments, a memory interface 504 can be a direct memory access (DMA) slave, to accommodate DMA transfers over a system memory initiated by a DMA master. Such a DMA master can be a device different from a host processor. In such configurations, processing module 500 can receive data for processing (e.g., DMA write), and transfer processed data out (e.g., DMA read) without consuming host processor resources.
  • Arbiter logic 506 can arbitrate between conflicting data accesses within processing module 500. In some embodiments, arbiter logic 506 can arbitrate between accesses by offload processor 508 and accesses external to the processor module 500. It is understood that a processing module 500 can include multiple locations that are operated on at the same time. It is also understood that accesses that are arbitrated by arbiter logic 506 can include accesses to physical system memory space occupied by the processor module 500, as well as accesses to resources (e.g., processor resources). Accordingly, arbitration rules for arbiter logic 506 can vary according to application. In some embodiments, such arbitration rules are fixed for a given processor module 500. In such cases, different applications can be accommodated by switching out different processing modules. However, in alternate embodiments, such arbitration rules can be configurable while the module is connected to a data bus.
  • Offload processor(s) 508 can include one or more processors that can operate on data transferred over the system memory bus. In some embodiments, offload processors can run a general operating system, enabling processor contexts to be saved and retrieved. Computing tasks executed by offload processor 508 can be controlled by control logic 512. Offload processor(s) 508 can operate on data buffered in the processing module 500. In addition or alternatively, offload processor(s) 508 can access data stored elsewhere in a system memory space. In some embodiment, offload processor(s) 508 can include a cache memory configured to store context information. An offload processor(s) 508 can include multiple cores or one core.
  • A processing module 500 can be included in a system having a host processor (not shown). In some embodiments, offload processors 508 can be a different type of processor as compared to the host processor. In particular embodiments, offload processors 508 can consume less power and/or have less computing power than a host processor. In very particular embodiments, offload processors 508 can be “wimpy” core processors, while a host processor can be a “brawny” core processor. In alternate embodiments, offload processors 508 can have equivalent or greater computing power than any host processor.
  • Local memory 510 can be connected to offload processor(s) 508 to enable the storing of context information. Accordingly, offload processor(s) 508 can store current context information, and then switch to a new computing task, then subsequently retrieve the context information to resume the prior task. In very particular embodiments, local memory 510 can be a low latency memory with respect to other memories in a system. In some embodiments, storing of context information can include copying a cache of an offload processor 508 to the local memory 510.
  • In some embodiments, a same space within local memory 510 is accessible by multiple offload processors 508 of the same type. In this way, a context stored by one offload processor can be resumed by a different offload processor.
  • Control logic 512 can control processing tasks executed by offload processor(s) 508. In some embodiments, control logic 512 can be considered a hardware scheduler that can be conceptualized as including a data evaluator 514, scheduler 516 and a switch controller 518. A data evaluator 514 can extract “metadata” from write data transferred over a system memory bus. “Metadata”, as used herein, can be any information embedded at one or more predetermined locations of a block of write data that indicates processing to be performed on all or a portion of the block of write data. In some embodiments, metadata can be data that indicates a higher level organization for the block of write data. As but one very particular embodiment, metadata can be header information of network packet (which may or may not be encapsulated within a higher layer packet structure).
  • A scheduler 516 can order computing tasks for offload processor(s) 508. In some embodiments, scheduler 516 can generate a schedule that is continually updated as write data for processing is received. In very particular embodiments, a scheduler 516 can generate such a schedule based on the ability to switch contexts of offload processor(s) 508. In this way, module computing priorities can be adjusted on the fly. In very particular embodiments, a scheduler 516 can assign a portion of physical address space to an offload processor 508, according to computing tasks. The offload processor 508 can then switch between such different spaces, saving context information prior to each switch, and subsequently restoring context information when returning to the memory space.
  • Switch controller 518 can control computing operations of offload processor(s) 508. In particular embodiments, according to scheduler 516, switch controller 518 can order offload processor(s) 510 to switch contexts. It is understood that a context switch operation can be an “atomic” operation, executed in response to a single command from switch controller 518. In addition or alternatively, a switch controller 518 can issue an instruction set that stores current context information, recalls context information, etc.
  • In some embodiments, processing module 500 can include a buffer memory (not shown). A buffer memory can store received write data on-board the processor module 500. A buffer memory can be implemented on an entirely different set of memory devices, or can be a memory embedded with logic and/or the offload processor. In the latter case, arbiter logic 506 can arbitrate access to the buffer memory. In some embodiments, a buffer memory can correspond to a portion of a system physical memory space. The remaining portion of the system memory space can correspond to other like processor modules and/or memory modules connected to the same system memory bus. In some embodiments buffer memory can be different than local memory 510. For example, buffer memory can have a slower access time than a local memory 510. However, in other embodiments, buffer memory and local memory can be implemented with like memory devices.
  • In very particular embodiments, write data for processing can have an expected maximum flow rate. A processor module 500 can be configured to operate on such data at, or faster than, such a flow rate. In this way, a master device (not shown) can write data to a processor module without danger of overwriting data “in process”.
  • The various computing elements of a processor module 500 can be implemented as one or more integrated circuit devices (ICs). It is understood that the various components shown in FIG. 5-1 can be formed in the same or different ICs. For example, control logic 512, memory interface 514, and/or arbiter logic 506 can be implemented on one or more logic ICs, while offload processor(s) 508 and local memory 510 are separate ICs. Logic ICs can be fixed logic (e.g., application specific ICs), programmable logic (e.g., field programmable gate arrays, FPGAs), or combinations thereof.
  • FIG. 5-2 shows a processor module 500-1 according to one very particular embodiment. A processor module 500-1 can include ICs 520-0/1 mounted to a printed circuit board (PCB) type substrate 522. PCB type substrate 522 can include in-line module connection 502, which in one very particular embodiment can be a DIMM compatible connection. IC 520-0 can be a system-on-chip (SoC) type device, integrating multiple functions. In the very particular embodiment shown, an IC 520-0 can include embedded processor(s), logic and memory. Such embedded processor(s) can be offload processor(s) 508 as described herein, or equivalents. Such logic can be any of controller logic 512, memory interface 504 and/or arbiter logic 506, as described herein, or equivalents. Such memory can be any of local memory 510, cache memory for offload processor(s) 508, or buffer memory, as described herein, or equivalents. Logic IC 520-1 can provide logic functions not included IC 520-0.
  • FIG. 5-3 shows a processor module 500-2 according to another very particular embodiment. A processor module 500-2 can include ICs 520-2, -3, -4, -5 mounted to a PCB type substrate 522, like that of FIG. 5-2. However, unlike FIG. 5-2, processor module functions are distributed among single purpose type ICs. IC 520-2 can be a processor IC, which can be an offload processor 508. IC 520-3 can be a memory IC which can include local memory 510, buffer memory, or combinations thereof. IC 520-4 can be a logic IC which can include control logic 512, and in one very particular embodiment, can be an FPGA. IC 520-5 can be another logic IC which can include memory interface 504 and arbiter logic 506, and in one very particular embodiment, can also be an FPGA.
  • It is understood that FIGS. 5-2 and 5-3 represent but two of various implementations. The various functions of a processor module can be distributed over any suitable number of ICs, including a single SoC type IC.
  • FIG. 5-4 shows an opposing side of a processor module 500-3 according to a very particular embodiment. Processor module 500-3 can include a number of memory ICs, one shown as 520-5, mounted to a PCB type substrate 522, like that of FIG. 5-2. It is understood that various processing and logic components can be mounted on an opposing side to that shown. Memory ICs 520-5 can be configured to represent a portion of the physical memory space of a system. Memory ICs 520-5 can perform any or all of the following functions: operate independently of other processor module components, providing system memory accessed in a conventional fashion; serve as buffer memory, storing write data that can be processed with other processor module components; or serve as local memory for storing processor context information.
  • FIG. 5-4 can also represent a conventional DIMM module (i.e., it serves only a memory function) that can populate a memory bus along with processor modules as described herein, or equivalents.
  • FIG. 5-5 shows a system 530 according to one embodiment. A system 530 can include a system memory bus 528 accessible via multiple in-line module slots (one shown as 526). According to embodiments, any or all of the slots 526 can be occupied by a processor module 500 as described herein, or an equivalent. In the event all slots 526 are not occupied by a processor module 500, available slots can be occupied by conventional in-line memory modules 524. In a very particular embodiment, slots 526 can be DIMM slots.
  • In some embodiments, a processor module 500 can occupy one slot. However, in other embodiments, a processor module can occupy multiple slots (i.e., include more than one connection). In some embodiments, a system memory bus 528 can be further interfaced with one or more host processors and/or input/output devices (not shown).
  • Having described processor modules according to various embodiments, operations of a processor module according to particular embodiments will now be described. FIGS. 5-6 to 5-11 show processor module operations according to various embodiments. FIGS. 5-10 to 5-15 show a processor module like that of FIG. 5-1, along with a system memory bus 528, and a buffer memory 532. It is understood that in some embodiments, a buffer memory 532 can part of processor module 500. In such a case, arbitration between accesses via system memory 528 and offload processors can be controlled by arbiter logic 506.
  • Referring to FIG. 5-6, write data 534-0 can be received on system memory bus 528 (circle “1”). In some embodiments, such an action can include the writing of data to a particular physical address space range of a system memory. In a very particular embodiment, such an action can be a DMA write independent of any host processor. Write data 534-0 can include metadata (MD) as well as data to be processed (Data). In the embodiment shown, write data 534-0 can correspond to a particular processing operation (Session0).
  • Control logic 512 can access metadata (MD) of the write data 534-0 to determine a type of processing to be performed (circle “2”). In some embodiments, such an action can include a direct read from a physical address (i.e., MD location is at a predetermined location). In addition or alternatively, such an action can be an indirect read (i.e., MD is accessed via pointer, or the like). The action shown by circle “2” can be performed by any of: a read by control logic 512 or read by an offload processor 508. From extracted metadata, scheduler 516 can create a processing schedule, or modify an existing schedule to accommodate the new computing task (circle “3”).
  • Referring to FIG. 5-7, in response to a scheduler 516, switch controller 518 can direct one or more offload processors 508 be begin processing data according to MD of the write data (circles “4”, “5”). Such processing of data can include any of the following and equivalents: offload processor 508 can process write data stored in a buffer memory of the processor module 500, with accesses being arbitrated by arbiter logic 506; offload processor 508 can operate on data previously received; offload processor 508 can receive and operation on data stored at a location different than the processor module 500.
  • Referring to FIG. 5-8, additional write data 534-1 can be received on system memory bus 528 (circle “6”). Write data 534-1 can include MD that indicates a different processing operation (Session1) than that for write data 534-0. Control logic 512 can access metadata (MD) of the new write data 534-1 to determine a type of processing to be performed (circle “7”). From extracted metadata, scheduler 516 can modify the current schedule to accommodate the new computing task (circle “8”). In the particular example shown, the modified schedule re-tasks offload processor(s) 508. Thus, switch controller 518 can direct an offload processor 508 to store its current context (ContextA) in local memory 510 (circle “9”).
  • Referring to FIG. 5-9, in response to switch controller 518, offload processor(s) 508 can begin the new processing task (circle “10”). Consequently, offload processor(s) 508 can maintain a new context (ContextB) corresponding to the new processing task.
  • Referring to FIG. 5-10, a processing task by offload processor 508 can be completed. In the very particular embodiment shown, such processing can modify write data (534-1) and such modified data 534-1′ can be read out over system memory bus 528 (circle “11”). In response to the completion of processing task, scheduler 516 can update a schedule. In the example shown, in response to the updated schedule, switch controller 518 can direct offload processor(s) 508 to restore the previously saved context (ContextA) from local memory 510 (circle “12”). As understood from above, in some particular embodiments, a restored context (e.g., ContextA) may have been stored by an offload processor different from the one that saved the context in the first place.
  • Referring to FIG. 5-11, with a previous context restored, offload processor(s) 508 can return to processing data according to the previous task (Session0) (circle “13”).
  • FIG. 5-12 shows a method 540 according an embodiment. A method 540 can include detecting the write of session data to a system memory with an in-line module slave interface 542. Such an action can include determining if received write data has metadata (i.e., data identifying a particular processing). It is understood that “session data” is data corresponding to a particular processing task. Further, it is understood that MD accompanying (or embedded within) session data can identify the priority of a session with respect to other sessions.
  • A method 540 can determine if current offload processing is sufficient for a new session or change of session 544. Such an action take into account a processing time required for any current sessions.
  • If current processing resources can accommodate new session requirements (Y from 544), a hardware schedule (schedule for controlling offload processor(s)) can be revised and the new session can be assigned to an offload processor. If current processing resources cannot accommodate new session requirements (N from 544), one or more offload processors can be selected for re-tasking (e.g., a context switch) 550 and the hardware schedule can be modified accordingly 552. The selected offload processors can save their current context data 554 and then switch to the new session 556.
  • FIG. 5-13 shows a method 560 according another embodiment. A method 560 can include determining if a computing session for an offload processor is complete 562 or has been terminated 564. In such cases (Y from 562/564), it can be determined if the free in-line module offload processor (i.e., an offload processor whose session is complete/terminated) has a stored context 566. That is, it can be determined if the free processor was previously operating on a session.
  • If a free offload processor was operating according to another session (Y from 566), the offload processor can restore the previous context 568. If a free offload processor has no stored context, it can be assigned to an existing session (if possible) 570. An existing hardware schedule can be updated correspondingly 572.
  • Processor modules according to embodiments herein can be employed to accomplish various processing tasks. According to some embodiments, processor modules can be attached to a system memory bus to operate on network packet data. Such embodiments will now be described.
  • FIG. 6-1 shows a system 601 that can transport packet data to one or more computational units (one shown as 600) located on a module, which in particular embodiments, can include a connector compatible with an existing memory module. In some embodiments, a computational unit 600 can include a processor module (e.g., XIMM) as described in embodiments herein, or an equivalent. A computational unit 600 can be capable of intercepting or otherwise accessing packets sent over a memory bus 616 and carrying out processing on such packets, including but not limited to termination or metadata processing. A system memory bus 616 can be a system memory bus like those described herein, or equivalents (e.g., 528).
  • According to some embodiments, packets corresponding to a particular flow can be transported to a storage location accessible by, or included within computational unit 600. Such transportation can occur without consuming resources of a host processor module 606 c connected to memory bus 616. In particular embodiments, such transport can occur without interrupting the host processor module 606 c. In such an arrangement, a host processor module 606 c does not have to handle incoming flows. Incoming flows can be directed to computational unit 600, which in particular embodiments can include one or more general purpose processors 608 i. Such general purpose processors 608 i can be capable of running code for terminating incoming flows.
  • In one very particular embodiment, a general purpose processor 608 i can run code for terminating particular network flow session types, such as Apache video sessions, as but one example.
  • In addition or alternatively, a general purpose processor 608 i can process metadata of a packet. In such embodiments, such metadata can include one or more fields of a header for the packet, or a header encapsulated further within the packet.
  • Referring still to FIG. 6-1, according to embodiments, a system 601 can carry out any of the following functions: 1) transport packets of a flow to a destination occupied by, or accessible by, a computational unit 600 without interrupting a host processor module 606 c; 2) transport packets to an offload processor 608 i capable of terminating session flows (i.e., the offload processor is responsible for terminating session flows); 3) transport packets to a midplane switch that can process the metadata associated with a packet and make a switching decision; or 4) provide a novel high speed packet terminating system.
  • Conventional packet processing systems can utilize host processors for packet termination. However, due to the context switching involved in handling multiple sessions, conventional approaches require significant processing overhead for such context switching, and can incur memory access and network stack delay.
  • In contrast to conventional approaches, embodiments as disclosed herein can enable high speed packet termination by reducing context switch overhead of a host processor. Embodiments can provide any of the following functions: 1) offload computation tasks to one or more processors via a system memory bus, without the knowledge of the host processor, or significant host processor involvement; 2) interconnect servers in a rack or amongst racks by employing offload processors as switches; or 3) use I/O virtualization to redirect incoming packets to different offload processors.
  • Referring still to FIG. 6-1, a system 601 can include an I/O device 602 which can receive packet or other I/O data from an external source. In some embodiments I/O device 602 can include physical or virtual functions generated by the physical device to receive a packet or other I/O data from the network or another computer or virtual machine. In the very particular embodiment shown, an I/O device 602 can include a network interface card (NIC) having input buffer 602 a (e.g., DMA ring buffer) and an I/O virtualization function 602 b.
  • According to embodiments, an I/O device 602 can write a descriptor including details of the necessary memory operation for the packet (i.e. read/write, source/destination). Such a descriptor can be assigned a virtual memory location (e.g., by an operating system of the system 601). I/O device 602 can communicate with an input output memory management unit (IOMMU) 604 which can translate virtual addresses to corresponding physical addresses. In the particular embodiment shown, a translation look-aside buffer (TLB) 604 a can be used for such translation. Virtual function reads or writes data between I/O device and system memory locations can then be executed with a direct memory transfer (e.g., DMA) via a memory controller 606 b of the system 601. An I/O device 602 can be connected to IOMMU 604 b by a host bus 612. In one very particular embodiment, a host bus 612 can be a peripheral interconnect (PCI) type bus. IOMMU 604 b can be connected to a host processing section 606 at a central processing unit I/O (CPUIO) 606 a. In the embodiment shown, such a connection 664 can support a HyperTransport (HT) protocol.
  • In the embodiment shown, a host processing section 606 can include the CPUIO 606 a, memory controller 606 b, host processor module 606 c and corresponding provisioning agent 606 d.
  • In particular embodiments, a computational unit 600 can interface with the system bus 616 via standard in-line module connection, which in very particular embodiments can include a DIMM type slot. In the embodiment shown, a memory bus 616 can be a DDR3 type memory bus, however alternate embodiments can include any suitable system memory bus. Packet data can be sent by memory controller 606 b to via memory bus 616 to a DMA slave interface 610 a. DMA slave interface 610 a can be adapted to receive encapsulated read/write instructions from a DMA write over the memory bus 616.
  • A hardware scheduler (608 b/c/d/e/h) can perform traffic management on incoming packets by categorizing them according to flow using session metadata. Packets can be queued for output in an onboard memory (610 b/608 a/608 m) based on session priority. When the hardware scheduler determines that a packet for a particular session is ready to be processed by the offload processor 608 i, the onboard memory is signaled for a context switch to that session. Utilizing this method of prioritization, context switching overhead can be reduced, as compared to conventional approaches. That is, a hardware scheduler can handle context switching decisions and thus optimizing the performance of the downstream resource (e.g., offload processor 608 i).
  • As noted above, in very particular embodiments, an offload processor 608 i can be a “wimpy” core type processor. According to some embodiments, a host processor module 606 c can include a “brawny” core type processor (e.g., an x86 or any other processor capable of handling “heavy touch” computational operations). While an I/O device 602 can be configured to trigger host processor interrupts in response to incoming packets, according to embodiments, such interrupts can be disabled, thereby reducing processing overhead for the host processor module 606 c. In some very particular embodiments, an offload processor 608 i can include an ARM, ARC, Tensilica, MIPS, Strong/ARM or any other processor capable of handling “light touch” operations. Preferably, an offload processor can run a general purpose operating system for executing a plurality of sessions, which can be optimized to work in conjunction with the hardware scheduler in order to reduce context switching overhead.
  • Referring still to FIG. 6-1, in operation, a system 601 can receive packets from an external network over a network interface. The packets can be directed for processing by either a host processor module 606 c or an offload processor 608 i based on the classification logic and schematics employed by I/O device 602. In particular embodiments, I/O device 602 can operate as a virtualized NIC, with packets for a particular logical network or to a certain virtual MAC (VMAC) address being directed into separate queues and sent over to the destination logical entity. Such an arrangement can transfer packets to different entities. In some embodiments, each such entity can have a virtual driver, a virtual device model that it uses to communicate with virtual network interfaces it is connected to.
  • According to embodiments, multiple devices can be used to redirect traffic to specific memory addresses. So, each of the network devices operates as if it is transferring the packets to the memory location of a logical entity. However, in reality, such packets can be transferred to memory addresses where they can be handled by one or more offload processors. In particular embodiments such transfers are to physical memory addresses, thus logical entities can be removed from the processing, and a host processor can be free from such packet handling.
  • Accordingly, embodiments can be conceptualized as providing a memory “black box” to which specific network data can be fed. Such a memory black box can handle the data (e.g., process it) and respond back when such data is requested.
  • Referring still to FIG. 6-1, according to some embodiments, I/O device 602 can receive data packets from a network or from a computing device. The data packets can have certain characteristics, including transport protocol number, source and destination port numbers, source and destination IP addresses, for example. The data packets can further have metadata that is processed (608 d) that helps in their classification and management.
  • I/O device 602 can include, but is not limited to, peripheral component interconnect (PCI) and/or PCI express (PCIe) devices connecting with host motherboard via PCI or PCIe bus (e.g., 612). Examples of I/O devices include a network interface controller (N IC), a host bus adapter, a converged network adapter, an ATM network interface etc.
  • In order to provide for an abstraction scheme that allows multiple logical entities to access the same I/O device 602, the I/O device may be virtualized to provide for multiple virtual devices each of which can perform some of the functions of the physical I/O device. The IO virtualization program, according to an embodiment, can redirect traffic to different memory locations (and thus to different offload processors attached to modules on a memory bus). To achieve this an I/O device 602 (e.g., a network card) may be partitioned into several function parts; including controlling function (CF) supporting input/output virtualization (IOV) architecture (e.g., single-root IOV) and multiple virtual function (VF) interfaces. Each virtual function interface may be provided with resources during runtime for dedicated usage. Examples of the CF and VF may include the physical function and virtual functions under schemes such as Single Root I/O Virtualization or Multi-Root I/O Virtualization architecture. The CF acts as the physical resources that sets up and manages virtual resources. The CF is also capable of acting as a full-fledged 10 device. The VF is responsible for providing an abstraction of a virtual device for communication with multiple logical entities/multiple memory regions.
  • The operating system/the hypervisor/any of the virtual machines/user code running on a host processor module 606 c may be loaded with a device model, a VF driver and a driver for a CF. The device model may be used to create an emulation of a physical device for the host processor 606 c to recognize each of the multiple VFs that are created. The device model may be replicated multiple times to give the impression to a VF driver (a driver that interacts with a virtual 10 device) that it is interacting with a physical device of a particular type.
  • For example, a certain device module may be used to emulate a network adapter such as the Intel® Ethernet Converged Network Adapter (CNA) X540-T2, so that the I/O device 602 believes it is interacting with such an adapter. In such a case, each of the virtual functions may have the capability to support the functions of the above said CNA, i.e., each of the Physical Functions should be able to support such functionality. The device model and the VF driver can be run in either privileged or non-privileged modes. In some embodiments, there is no restriction with regard to who hosts/runs the code corresponding to the device model and the VF driver. The code, however, has the capability to create multiple copies of device model and VF driver so as to enable multiple copies of said I/O interface to be created.
  • An application or provisioning agent 606 d, as part of an application/user level code running in a kernel, may create a virtual I/O address space for each VF during runtime and allocate part of the physical address space to it. For example, if an application handling the VF driver instructs it to read or write packets from or to memory addresses 0xaaaa to 0xffff, the device driver may write I/O descriptors into a descriptor queue with a head and tail pointer that are changed dynamically as queue entries are filled. The data structure may be of another type as well, including but not limited to a ring structure 602 a or hash table.
  • The VF can read from or write data to the address location pointed to by the driver (and hence to a computational unit 600). Further, on completing the transfer of data to the address space allocated to the driver, interrupts, which are usually triggered to the host processor to handle said network packets, can be disabled. Allocating a specific I/O space to a device can include allocating said IO space a specific physical memory space occupied.
  • In another embodiment, the descriptor may comprise only a write operation, if the descriptor is associated with a specific data structure for handling incoming packets. Further, the descriptor for each of the entries in the incoming data structure may be constant so as to redirect all data write to a specific memory location. In an alternate embodiment, the descriptor for consecutive entries may point to consecutive entries in memory so as to direct incoming packets to consecutive memory locations.
  • Alternatively, said operating system may create a defined physical address space for an application supporting the VF drivers and allocate a virtual memory address space to the application or provisioning agent 606 d, thereby creating a mapping for each virtual function between said virtual address and a physical address space. Said mapping between virtual memory address space and physical memory space may be stored in IOMMU tables 604 a. The application performing memory reads or writes may supply virtual addresses to say virtual function, and the host processor OS may allocate a specific part of the physical memory location to such an application.
  • Alternatively, VF may be configured to generate requests such as read and write which may be part of a direct memory access (DMA) read or write operation, for example. The virtual addresses is be translated by the IOMMU 604 b to their corresponding physical addresses and the physical addresses may be provided to the memory controller for access. That is, the IOMMU 604 b may modify the memory requests sourced by the I/O devices to change the virtual address in the request to a physical address, and the memory request may be forwarded to the memory controller for memory access. The memory request may be forwarded over a bus 614. The VF may in such cases carry out a direct memory access by supplying the virtual memory address to the IOMMU 604 b.
  • Alternatively, said application may directly code the physical address into the VF descriptors if the VF allows for it. If the VF cannot support physical addresses of the form used by the host processor 606 c, an aperture with a hardware size supported by the VF device may be coded into the descriptor so that the VF is informed of the target hardware address of the device. Data that is transferred to an aperture may be mapped by a translation table to a defined physical address space in the system memory. The DMA operations may be initiated by software executed by the processors, programming the I/O devices directly or indirectly to perform the DMA operations.
  • Referring still to FIG. 6-1, in particular embodiments, parts of computational unit 600 can be implemented with one or more FPGAs. In the system of FIG. 6-1, computational unit 600 can include FPGA 610 in which can be formed a DMA slave device module 610 a and arbiter 610 f. A DMA slave module 610 a can be any device suitable for attachment to a memory bus 616 that can respond to DMA read/write requests. In alternate embodiments, a DMA slave module 610 a can be another interface capable of block data transfers over memory bus 616. The DMA slave module 610 a can be capable of receiving data from a DMA controller (when it performs a read from a ‘memory’ or from a peripheral) or transferring data to a DMA controller (when it performs a write instruction on the DMA slave module 610 a). The DMA slave module 610 a may be adapted to receive DMA read and write instructions encapsulated over a memory bus, (e.g., in the form of a DDR data transmission, such as a packet or data burst), or any other format that can be sent over the corresponding memory bus.
  • A DMA slave module 610 a can reconstruct the DMA read/write instruction from the memory R/W packet. The DMA slave module 610 a may be adapted to respond to these instructions in the form of data reads/data writes to a DMA master, which could either be housed in a peripheral device, in the case of a PCIe bus, or a system DMA controller in the case of an ISA bus.
  • I/O data that is received by the DMA device 610 a can then be queued for arbitration. Arbitration is the process of scheduling packets of different flows, such that they are provided access to available bandwidth based on a number of parameters. In general, an arbiter provides resource access to one or more requestors. If multiple requestors request access, an arbiter 610 f can determine which requestor becomes the accessor and then passes data from the accessor to the resource interface, and the downstream resource can begin execution on the data. After the data has been completely transferred to a resource, and the resource has competed execution, the arbiter 610 f can transfer control to a different requestor and this cycle repeats for all available requestors. In the embodiment of FIG. 6-1, arbiter 610 f can notify other portions of computational unit 600 (e.g., 608) of incoming data.
  • Alternatively, a computation unit 600 can utilize an arbitration scheme shown in U.S. Pat. No. 7,863,283, issued to Dalal on Oct. 62, 2010, the content of which are incorporated herein by reference. Other suitable arbitration schemes known in art could be implemented in embodiments herein. Alternatively, the arbitration scheme for an embodiment can be an OpenFlow switch and an OpenFlow controller.
  • In the very particular embodiment of FIG. 6-1, computational unit 600 can further include notify/prefetch circuits 610 c which can prefetch data stored in a buffer memory 610 b in response to DMA slave module 610 a, and as arbitrated by arbiter 610 f. Further, arbiter 610 f can access other portions of the computational unit 600 via a memory mapped I/O ingress path 610 e and egress path 610 g.
  • Referring to FIG. 6-1, a hardware scheduler can include a scheduling circuit 608 b/n to implement traffic management of incoming packets. Packets from a certain source, relating to a certain traffic class, pertaining to a specific application or flowing to a certain socket are referred to as part of a session flow and are classified using session metadata. Such classification can be performed by classifier 608 e.
  • In some embodiments, session metadata 608 d can serve as the criterion by which packets are prioritized and scheduled and as such, incoming packets can be reordered based on their session metadata. This reordering of packets can occur in one or more buffers and can modify the traffic shape of these flows. The scheduling discipline chosen for this prioritization, or traffic management (TM), can affect the traffic shape of flows and micro-flows through delay (buffering), bursting of traffic (buffering and bursting), smoothing of traffic (buffering and rate-limiting flows), dropping traffic (choosing data to discard so as to avoid exhausting the buffer), delay jitter (temporally shifting cells of a flow by different amounts) and by not admitting a connection (e.g., cannot simultaneously guarantee existing service (SLAB) with an additional flow's SLA).
  • According to embodiments, computational unit 600 can serve as part of a switch fabric, and provide traffic management with depth-limited output queues, the access to which is arbitrated by a scheduling circuit 608 b/n. Such output queues are managed using a scheduling discipline to provide traffic management for incoming flows. The session flows queued in each of these queues can be sent out through an output port to a downstream network element.
  • It is noted that some conventional traffic management circuits do not take into the account the handling and management of data by downstream elements except for meeting the SLA agreements it already has with said downstream elements. In contrast, according to embodiments, a scheduler circuit 608 b/n can allocate a priority to each of the output queues and carry out reordering of incoming packets to maintain persistence of session flows in these queues. A scheduler circuit 608 b/n can be used to control the scheduling of each of these persistent sessions into a general purpose operating system (OS) 608 j, executed on an offload processor 608 i. Packets of a particular session flow, as defined above, can belong to a particular queue. The scheduler circuit 608 b/n may control the prioritization of these queues such that they are arbitrated for handling by a general purpose (GP) processing resource (e.g., offload processor 608 i) located downstream. An OS 608 j running on a downstream processor 608 i can allocate execution resources such as processor cycles and memory to a particular queue it is currently handling. The OS 608 j may further allocate a thread or a group of threads for that particular queue, so that it is handled distinctly by the general purpose processing element 608 i as a separate entity. Thus, in some embodiments there can be multiple sessions running on a GP processing resource, each handling data from a particular session flow resident in a queue established by the scheduler circuit, to tightly integrate the scheduler and the downstream resource (e.g., 608 i). This can bring about persistence of session information across the traffic management and scheduling circuit and the general purpose processing resource 608 j.
  • Dedicated computing resources (e.g., 608 i), memory space and session context information for each of the sessions can provide a way of handling, processing and/or terminating each of the session flows at the general purpose processor 608 i. The scheduler circuit 608 b/n can exploit this functionality of the execution resource to queue session flows for scheduling downstream. For example, the scheduler circuit 608 b/n can be informed of the state of the execution resource(s) (e.g., 608 i), the current session that is run on the execution resource; the memory space allocated to it, the location of the session context in the processor cache.
  • According to embodiments, a scheduler circuit 608 b/n can further include switching circuits to change execution resources from one state to another. The scheduler circuit 608 b/n can use such a capability to arbitrate between the queues that are ready to be switched into the downstream execution resource. Further, the downstream execution resource can be optimized to reduce the penalty and overhead associated with context switch between resources. This is further exploited by the scheduler circuit 608 b/n to carry out seamless switching between queues, and consequently their execution as different sessions by the execution resource.
  • A scheduler circuit 608 b/n according to embodiments can schedule different sessions on a downstream processing resource, wherein the two are operated in coordination to reduce the overhead during context switches. An important factor to decreasing the latency of services and engineering computational availability can be hardware context switching synchronized with network queuing. In embodiments, when a queue is selected by a traffic manager, a pipeline coordinates swapping in of the cache (e.g., L2 cache) of the corresponding resource and transfers the reassembled I/O data into the memory space of the executing process. In certain cases, no packets are pending in the queue, but computation is still pending to service previous packets. Once this process makes a memory reference outside of the data swapped, the scheduler circuit can enable queued data from an I/O device 602 to continue scheduling the thread.
  • In some embodiments, to provide fair queuing to a process not having data, a maximum context size can be assumed as data processed. In this way, a queue can be provisioned as the greater of computational resource and network bandwidth resource. As but one very particular example, a computation resource can be an ARM A9 processor running at 800 MHz, while a network bandwidth can be 3 Gbps of bandwidth. Given the lopsided nature of this ratio, embodiments can utilize computation having many parallel sessions (such that the hardware's prefetching of session-specific data offloads a large portion of the host processor load) and having minimal general purpose processing of data.
  • Accordingly, in some embodiments, a scheduler circuit 608 b/n can be conceptualized as arbitrating, not between outgoing queues at line rate speeds, but arbitrating between terminated sessions at very high speeds. The stickiness of sessions across a pipeline of stages, including a general purpose OS, can be a scheduler circuit optimizing any or all such stages of such a pipeline.
  • Alternatively, a scheduling scheme can be used as shown in U.S. Pat. No. 7,760,765 issued to Dalal on Jul. 20, 2010, incorporated herein by reference. This scheme can be useful when it is desirable to rate limit the flows for preventing the downstream congestion of another resource specific to the over-selected flow, or for enforcing service contracts for particular flows. Embodiments can include arbitration scheme that allows for service contracts of downstream resources, such as general purpose OS that can be enforced seamlessly.
  • Referring still to FIG. 6-1, a hardware scheduler according to embodiments herein, or equivalents, can provide for the classification of incoming packet data into session flows based on session metadata. It can further provide for traffic management of these flows before they are arbitrated and queued as distinct processing entities on the offload processors.
  • In some embodiments, offload processors (e.g., 608 i) can be general purpose processing units capable of handling packets of different application or transport sessions. Such offload processors can be low power processors capable of executing general purpose instructions. The offload processors could be any suitable processor, including but not limited to: ARM, ARC, Tensilica, MIPS, StrongARM or any other processor that serves the functions described herein. The offload processors have general purpose OS running on them, wherein the general purpose OS is optimized to reduce the penalty associated with context switching between different threads or group of threads.
  • In contrast, context switches on host processors can be computationally intensive processes that require the register save area, process context in the cache and TLB entries to be restored if they are invalidated or overwritten. Instruction Cache misses in host processing systems can lead to pipeline stalls and data cache misses lead to operation stall and such cache misses reduce processor efficiency and increase processor overhead.
  • According to embodiments, an OS 608 j running on the offload processors 608 i in association with a scheduler circuit, can operate together to reduce the context switch overhead incurred between different processing entities running on it. Embodiments can include a cooperative mechanism between a scheduler circuit and the OS on the offload processor 608 i, wherein the OS sets up session context to be physically contiguous (physically colored allocator for session heap and stack) in the cache; then communicates the session color, size, and starting physical address to the scheduler circuit upon session initialization. During an actual context switch, a scheduler circuit can identify the session context in the cache by using these parameters and initiate a bulk transfer of these contents to an external low latency memory. In addition, the scheduler circuit can manage the prefetch of the old session if its context was saved to a local memory 608 g. In particular embodiments, a local memory 608 g can be low latency memory, such as a reduced latency dynamic random access memory (RLDRAM), as but one very particular embodiment. Thus, in embodiments, session context can be identified distinctly in the cache.
  • In some embodiments, context size can be limited to ensure fast switching speeds. In addition or alternatively, embodiments can include a bulk transfer mechanism to transfer out session context to a local memory 608 g. The cache contents stored therein can then be retrieved and prefetched during context switch back to a previous session. Different context session data can be tagged and/or identified within the local memory 608 m for fast retrieval. As noted above, context stored by one offload processor may be recalled by a different offload processor.
  • In the very particular embodiment of FIG. 6-1 multiple offload processing cores can be integrated into a computation FPGA 608. Multiple computational FPGAs can be arbitrated by arbitrator circuits in another FPGA 610. The combination of computational FPGAs (e.g., 608) and arbiter FPGAs (e.g., 610) can be one implementation of a XIMM module. In particular applications, these XIMM modules can provide integrated traffic and thread management circuits that broker execution of multiple sessions on the offload processors.
  • FIG. 6-2 shows a system flow according to an embodiment. Packet or other I/O data can be received at an I/O device 620. An I/O device 620 can be physical device, virtual device or combination thereof. Interrupts generated from the I/O data, that would conventionally be intended for a host processor 624, can be disabled, allowing such I/O data to be processed without resources of the host processor 624.
  • An IOMMU 621 can map received data to physical addresses of a system address space. DMA master 625 can transmit such data to such memory addresses by operation of a memory controller 622. Memory controller 622 can execute DRAM transfers over a memory bus with a DMA Slave 627. Upon receiving transferred I/O data, a hardware scheduler 623 can schedule processing of such data with an offload processor 624. In some embodiments, a type of processing can be indicated by metadata within the I/O data. Further, in some embodiments such data can be stored in an Onboard Memory 629. According to instructions from hardware scheduler 623, one or more offload processors 624 can execute computing functions in response to the I/O data. In some embodiments, such computing functions can operate on the I/O data, and such data can be subsequently read out on memory bus via a read request processed by DMA Slave 627.
  • It should be appreciated that in the foregoing description of exemplary embodiments of the invention, various features of the invention are sometimes grouped together in a single embodiment, figure, or description thereof for the purpose of streamlining the disclosure aiding in the understanding of one or more of the various inventive aspects. This method of disclosure, however, is not to be interpreted as reflecting an intention that the claimed invention requires more features than are expressly recited in each claim. Rather, as the following claims reflect, inventive aspects lie in less than all features of a single foregoing disclosed embodiment. Thus, the claims following the detailed description are hereby expressly incorporated into this detailed description, with each claim standing on its own as a separate embodiment of this invention.
  • It is also understood that the embodiments of the invention may be practiced in the absence of an element and/or step not specifically disclosed. That is, an inventive feature of the invention may be elimination of an element.
  • Accordingly, while the various aspects of the particular embodiments set forth herein have been described in detail, the present invention could be subject to various changes, substitutions, and alterations without departing from the spirit and scope of the invention.

Claims (10)

    What is claimed is:
  1. 1. A method of processing structured data, comprising the steps of:
    providing a plurality of XIMM modules connected to a memory bus in a first server, with the XIMM modules each respectively having a DMA slave module connected to the memory bus and an arbiter for scheduling tasks, with the XIMM modules further providing an in-memory database; and
    connecting a central processing unit (CPU) in the first server to the XIMM modules by the memory bus, with the CPU arranged to process and direct structured queries to the plurality of XIMM modules.
  2. 2. The method of processing structured data of claim 1, wherein the XIMM modules communicate with each other without requiring access to a processor of the first server.
  3. 3. The method of processing structured data of claim 1, wherein the XIMM modules are mounted on different servers in the same rack, and further comprising a top of the rack switch to mediate communication therebetween.
  4. 4. The method of processing structured data of claim 1, wherein a XIMM driver executes a mmap routine to transfer a query from the CPU to the XIMM in the form of memory reads/writes.
  5. 5. The method of processing structured data claim 1, wherein the XIMM module is configured for insertion into a DIMM socket, and the XIMM module further comprises offload processors connected to memory and a computational FPGA.
  6. 6. A method of processing unstructured data, comprising the steps of:
    providing a plurality of XIMM modules connected to a memory bus, with the XIMM modules each respectively having a DMA slave module connected to the memory bus and an arbiter for scheduling tasks, with the XIMM modules providing an in-memory database; and
    connecting a central processing unit (CPU) to the XIMM modules by the memory bus, with the CPU arranged to process computationally intensive tasks data processing tasks while directing parallel computation tasks to the plurality of XIMM modules.
  7. 7. The method of processing unstructured data of claim 6, wherein a Map/Reduce algorithm is processed, with results of a Map step being stored and available in main memory, and collected through DMA operations and parsed by the XIMM modules.
  8. 8. The method of processing unstructured data of claim 6, wherein the XIMM modules together define a massively parallel I/O mid-plane defined by the XIMMs.
  9. 9. The method of processing unstructured data of claim 6, wherein a Map/Reduce algorithm is processed by multiple servers supporting XIMM modules, and intermediate (key, value) pairs are exchanged across the multiple servers through intelligent virtual switching of the XIMM modules.
  10. 10. The method of processing unstructured data of claim 6, wherein the XIMM modules are configured for insertion into a DIMM socket, and the XIMM module further comprises offload processors connected to memory and a computational FPGA.
US13900333 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors Abandoned US20130318269A1 (en)

Priority Applications (11)

Application Number Priority Date Filing Date Title
US201261650373 true 2012-05-22 2012-05-22
US201361753904 true 2013-01-17 2013-01-17
US201361753906 true 2013-01-17 2013-01-17
US201361753892 true 2013-01-17 2013-01-17
US201361753910 true 2013-01-17 2013-01-17
US201361753895 true 2013-01-17 2013-01-17
US201361753901 true 2013-01-17 2013-01-17
US201361753899 true 2013-01-17 2013-01-17
US201361753907 true 2013-01-17 2013-01-17
US201361753903 true 2013-01-17 2013-01-17
US13900333 US20130318269A1 (en) 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US13900333 US20130318269A1 (en) 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors

Publications (1)

Publication Number Publication Date
US20130318269A1 true true US20130318269A1 (en) 2013-11-28

Family

ID=49622482

Family Applications (8)

Application Number Title Priority Date Filing Date
US13900359 Active US9286472B2 (en) 2012-05-22 2013-05-22 Efficient packet handling, redirection, and inspection using offload processors
US13900318 Active 2034-01-03 US9558351B2 (en) 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors
US13900333 Abandoned US20130318269A1 (en) 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors
US13900273 Abandoned US20130318280A1 (en) 2012-05-22 2013-05-22 Offloading of computation for rack level servers and corresponding methods and systems
US13900262 Abandoned US20130318268A1 (en) 2012-05-22 2013-05-22 Offloading of computation for rack level servers and corresponding methods and systems
US13900367 Abandoned US20140165196A1 (en) 2012-05-22 2013-05-22 Efficient packet handling, redirection, and inspection using offload processors
US15396330 Pending US20170235699A1 (en) 2012-05-22 2016-12-30 Architectures and methods for processing data in parallel using offload processing modules insertable into servers
US15396318 Pending US20170237672A1 (en) 2012-05-22 2016-12-30 Network server systems, architectures, components and related methods

Family Applications Before (2)

Application Number Title Priority Date Filing Date
US13900359 Active US9286472B2 (en) 2012-05-22 2013-05-22 Efficient packet handling, redirection, and inspection using offload processors
US13900318 Active 2034-01-03 US9558351B2 (en) 2012-05-22 2013-05-22 Processing structured and unstructured data using offload processors

Family Applications After (5)

Application Number Title Priority Date Filing Date
US13900273 Abandoned US20130318280A1 (en) 2012-05-22 2013-05-22 Offloading of computation for rack level servers and corresponding methods and systems
US13900262 Abandoned US20130318268A1 (en) 2012-05-22 2013-05-22 Offloading of computation for rack level servers and corresponding methods and systems
US13900367 Abandoned US20140165196A1 (en) 2012-05-22 2013-05-22 Efficient packet handling, redirection, and inspection using offload processors
US15396330 Pending US20170235699A1 (en) 2012-05-22 2016-12-30 Architectures and methods for processing data in parallel using offload processing modules insertable into servers
US15396318 Pending US20170237672A1 (en) 2012-05-22 2016-12-30 Network server systems, architectures, components and related methods

Country Status (1)

Country Link
US (8) US9286472B2 (en)

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9262225B2 (en) 2009-10-30 2016-02-16 Iii Holdings 2, Llc Remote memory access functionality in a cluster of data processing nodes
US9311269B2 (en) 2009-10-30 2016-04-12 Iii Holdings 2, Llc Network proxy for high-performance, low-power data center interconnect fabric
US9465771B2 (en) 2009-09-24 2016-10-11 Iii Holdings 2, Llc Server on a chip and node cards comprising one or more of same
US9479463B2 (en) 2009-10-30 2016-10-25 Iii Holdings 2, Llc System and method for data center security enhancements leveraging managed server SOCs
US9509552B2 (en) 2009-10-30 2016-11-29 Iii Holdings 2, Llc System and method for data center security enhancements leveraging server SOCs or server fabrics
US9585281B2 (en) 2011-10-28 2017-02-28 Iii Holdings 2, Llc System and method for flexible storage and networking provisioning in large scalable processor installations
US9648102B1 (en) * 2012-12-27 2017-05-09 Iii Holdings 2, Llc Memcached server functionality in a cluster of data processing nodes
US9680770B2 (en) 2009-10-30 2017-06-13 Iii Holdings 2, Llc System and method for using a multi-protocol fabric module across a distributed server interconnect fabric
US9792249B2 (en) 2011-10-31 2017-10-17 Iii Holdings 2, Llc Node card utilizing a same connector to communicate pluralities of signals
US9876735B2 (en) 2009-10-30 2018-01-23 Iii Holdings 2, Llc Performance and power optimized computer system architectures and methods leveraging power optimized tree fabric interconnect
US9934395B2 (en) 2015-09-11 2018-04-03 International Business Machines Corporation Enabling secure big data analytics in the cloud

Families Citing this family (29)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP5939123B2 (en) * 2012-10-09 2016-06-22 富士通株式会社 Execution control program execution control method and an information processing apparatus
US9448837B2 (en) * 2012-12-27 2016-09-20 Nvidia Corporation Cooperative thread array granularity context switch during trap handling
US20160205071A1 (en) * 2013-09-23 2016-07-14 Mcafee, Inc. Providing a fast path between two entities
KR20150033453A (en) * 2013-09-24 2015-04-01 주식회사 엘지씨엔에스 Method of big data processing, apparatus performing the same and storage media storing the same
US9811467B2 (en) * 2014-02-03 2017-11-07 Cavium, Inc. Method and an apparatus for pre-fetching and processing work for procesor cores in a network processor
US20150254322A1 (en) * 2014-03-05 2015-09-10 International Business Machines Corporation Performing data analytics utilizing a user configurable group of reusable modules
CN103905337B (en) * 2014-03-31 2018-01-23 华为技术有限公司 A network resource processing apparatus, a method and system
US9383989B1 (en) 2014-06-16 2016-07-05 Symantec Corporation Systems and methods for updating applications
US20160026605A1 (en) * 2014-07-28 2016-01-28 Emulex Corporation Registrationless transmit onload rdma
US20160034291A1 (en) * 2014-07-29 2016-02-04 Freescale Semiconductor, Inc. System on a chip and method for a controller supported virtual machine monitor
US20160034205A1 (en) * 2014-08-01 2016-02-04 Software Ag Usa, Inc. Systems and/or methods for leveraging in-memory storage in connection with the shuffle phase of mapreduce
US9697114B2 (en) * 2014-08-17 2017-07-04 Mikhael Lerman Netmory
US9397952B2 (en) * 2014-09-05 2016-07-19 Futurewei Technologies, Inc. Segment based switching architecture with hybrid control in SDN
US9858104B2 (en) * 2014-09-24 2018-01-02 Pluribus Networks, Inc. Connecting fabrics via switch-to-switch tunneling transparent to network servers
US20160094630A1 (en) * 2014-09-30 2016-03-31 International Business Machines Corporation Mapping and reducing
US9762508B2 (en) 2014-10-02 2017-09-12 Microsoft Technology Licensing, Llc Relay optimization using software defined networking
US20160210077A1 (en) * 2015-01-20 2016-07-21 Ultrata Llc Trans-cloud object based memory
US9971506B2 (en) 2015-01-20 2018-05-15 Ultrata, Llc Distributed index for fault tolerant object memory fabric
US9886210B2 (en) * 2015-06-09 2018-02-06 Ultrata, Llc Infinite memory fabric hardware implementation with router
US9971542B2 (en) 2015-06-09 2018-05-15 Ultrata, Llc Infinite memory fabric streams and APIs
WO2016118624A1 (en) * 2015-01-20 2016-07-28 Ultrata Llc Object memory instruction set
US20180011806A1 (en) * 2015-01-28 2018-01-11 Hewlett-Packard Development Company, L.P. Supporting different types of memory devices
US20160292117A1 (en) * 2015-03-30 2016-10-06 Integrated Device Technology, Inc. Methods and Apparatus for Efficient Network Analytics and Computing Card
EP3089035A1 (en) * 2015-04-30 2016-11-02 Virtual Open Systems Virtualization manager for reconfigurable hardware accelerators
US9959306B2 (en) * 2015-06-12 2018-05-01 International Business Machines Corporation Partition-based index management in hadoop-like data stores
US9667657B2 (en) * 2015-08-04 2017-05-30 AO Kaspersky Lab System and method of utilizing a dedicated computer security service
US20170212783A1 (en) * 2016-01-22 2017-07-27 Samsung Electronics Co., Ltd. Electronic system with data exchange mechanism and method of operation thereof
US9984009B2 (en) * 2016-01-28 2018-05-29 Silicon Laboratories Inc. Dynamic containerized system memory protection for low-energy MCUs
US20170302438A1 (en) * 2016-04-15 2017-10-19 The Florida International University Board Of Trustees Advanced bus architecture for aes-encrypted high-performance internet-of-things (iot) embedded systems

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20070294433A1 (en) * 2006-06-14 2007-12-20 Leigh Kevin B system of implementing switch devices in a server system
US20080259555A1 (en) * 2006-01-13 2008-10-23 Sun Microsystems, Inc. Modular blade server
US20100031253A1 (en) * 2008-07-29 2010-02-04 Electronic Data Systems Corporation System and method for a virtualization infrastructure management environment
US20100235484A1 (en) * 2009-03-12 2010-09-16 International Business Machines Corporation Remotely Administering A Server
US20110035540A1 (en) * 2009-08-10 2011-02-10 Adtron, Inc. Flash blade system architecture and method
US20120047126A1 (en) * 2007-08-27 2012-02-23 Teradata Us, Inc. Methods and systems for hardware acceleration of streamed database operations and queries based on multiple hardware accelerators
US20120151292A1 (en) * 2010-12-14 2012-06-14 Microsoft Corporation Supporting Distributed Key-Based Processes

Family Cites Families (106)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5446844A (en) * 1987-10-05 1995-08-29 Unisys Corporation Peripheral memory interface controller as a cache for a large data processing system
US5237662A (en) 1991-06-27 1993-08-17 Digital Equipment Corporation System and method with a procedure oriented input/output mechanism
US5247675A (en) 1991-08-09 1993-09-21 International Business Machines Corporation Preemptive and non-preemptive scheduling and execution of program threads in a multitasking operating system
US5577213A (en) 1994-06-03 1996-11-19 At&T Global Information Solutions Company Multi-device adapter card for computer
US6085307A (en) 1996-11-27 2000-07-04 Vlsi Technology, Inc. Multiple native instruction set master/slave processor arrangement and method thereof
US20060117274A1 (en) 1998-08-31 2006-06-01 Tseng Ping-Sheng Behavior processor system and method
US5870350A (en) 1997-05-21 1999-02-09 International Business Machines Corporation High performance, high bandwidth memory bus architecture utilizing SDRAMs
US6092146A (en) 1997-07-31 2000-07-18 Ibm Dynamically configurable memory adapter using electronic presence detects
US6157989A (en) * 1998-06-03 2000-12-05 Motorola, Inc. Dynamic bus arbitration priority and task switching based on shared memory fullness in a multi-processor system
US6157955A (en) * 1998-06-15 2000-12-05 Intel Corporation Packet processing system including a policy engine having a classification unit
US6578110B1 (en) * 1999-01-21 2003-06-10 Sony Computer Entertainment, Inc. High-speed processor system and cache memories with processing capabilities
US6625685B1 (en) 2000-09-20 2003-09-23 Broadcom Corporation Memory controller with programmable configuration
US7120155B2 (en) 2000-10-03 2006-10-10 Broadcom Corporation Switch having virtual shared memory
KR100483643B1 (en) 2001-06-13 2005-04-18 가부시키가이샤 히타치세이사쿠쇼 Memory device
US6751113B2 (en) 2002-03-07 2004-06-15 Netlist, Inc. Arrangement of integrated circuits in a memory module
US7472205B2 (en) * 2002-04-24 2008-12-30 Nec Corporation Communication control apparatus which has descriptor cache controller that builds list of descriptors
EP1546929A4 (en) 2002-09-18 2008-07-02 Netezza Corp Intelligent storage device controller
US7454749B2 (en) 2002-11-12 2008-11-18 Engineered Intelligence Corporation Scalable parallel processing on shared memory computers
US7089412B2 (en) 2003-01-17 2006-08-08 Wintec Industries, Inc. Adaptive memory module
US7155379B2 (en) 2003-02-25 2006-12-26 Microsoft Corporation Simulation of a PCI device's memory-mapped I/O registers
US7337314B2 (en) 2003-04-12 2008-02-26 Cavium Networks, Inc. Apparatus and method for allocating resources within a security processor
US6982892B2 (en) 2003-05-08 2006-01-03 Micron Technology, Inc. Apparatus and methods for a physical layout of simultaneously sub-accessible memory modules
US20050038946A1 (en) * 2003-08-12 2005-02-17 Tadpole Computer, Inc. System and method using a high speed interface in a system having co-processors
US8776050B2 (en) 2003-08-20 2014-07-08 Oracle International Corporation Distributed virtual machine monitor for managing multiple virtual resources across multiple physical nodes
US7657706B2 (en) 2003-12-18 2010-02-02 Cisco Technology, Inc. High speed memory and input/output processor subsystem for efficiently allocating and using high-speed memory and slower-speed memory
US20050018495A1 (en) 2004-01-29 2005-01-27 Netlist, Inc. Arrangement of integrated circuits in a memory module
US9128632B2 (en) 2009-07-16 2015-09-08 Netlist, Inc. Memory module with distributed data buffers and method of operation
US7289386B2 (en) 2004-03-05 2007-10-30 Netlist, Inc. Memory module decoder
US7916574B1 (en) 2004-03-05 2011-03-29 Netlist, Inc. Circuit providing load isolation and memory domain translation for memory module
US7286436B2 (en) 2004-03-05 2007-10-23 Netlist, Inc. High-density memory module utilizing low-density memory components
US7532537B2 (en) 2004-03-05 2009-05-12 Netlist, Inc. Memory module with a circuit providing load isolation and memory domain translation
US7668165B2 (en) 2004-03-31 2010-02-23 Intel Corporation Hardware-based multi-threading for packet processing
US7254036B2 (en) 2004-04-09 2007-08-07 Netlist, Inc. High density memory module using stacked printed circuit boards
US7480611B2 (en) 2004-05-13 2009-01-20 International Business Machines Corporation Method and apparatus to increase the usable memory capacity of a logic simulation hardware emulator/accelerator
US20060004965A1 (en) 2004-06-30 2006-01-05 Tu Steven J Direct processor cache access within a system having a coherent multi-processor protocol
US7305574B2 (en) 2004-10-29 2007-12-04 International Business Machines Corporation System, method and storage medium for bus calibration in a memory subsystem
KR100666169B1 (en) 2004-12-17 2007-01-09 삼성전자주식회사 Flash memory data storing device
US7380038B2 (en) * 2005-02-04 2008-05-27 Microsoft Corporation Priority registers for biasing access to shared resources
KR101375763B1 (en) 2005-04-21 2014-03-19 바이올린 메모리 인코포레이티드 Interconnection System
WO2007006146A1 (en) * 2005-07-12 2007-01-18 Advancedio Systems Inc. System and method of offloading protocol functions
US7442050B1 (en) 2005-08-29 2008-10-28 Netlist, Inc. Circuit card with flexible connection for memory module with heat spreader
US7650557B2 (en) * 2005-09-19 2010-01-19 Network Appliance, Inc. Memory scrubbing of expanded memory
US8225297B2 (en) 2005-12-07 2012-07-17 Microsoft Corporation Cache metadata identifiers for isolation and sharing
US7904688B1 (en) 2005-12-21 2011-03-08 Trend Micro Inc Memory management unit for field programmable gate array boards
US7619893B1 (en) 2006-02-17 2009-11-17 Netlist, Inc. Heat spreader for electronic modules
US20070226745A1 (en) 2006-02-28 2007-09-27 International Business Machines Corporation Method and system for processing a service request
US7421552B2 (en) 2006-03-17 2008-09-02 Emc Corporation Techniques for managing data within a data storage system utilizing a flash-based memory vault
US7434002B1 (en) 2006-04-24 2008-10-07 Vmware, Inc. Utilizing cache information to manage memory access and cache utilization
US7716411B2 (en) 2006-06-07 2010-05-11 Microsoft Corporation Hybrid memory device with single interface
WO2007147170A3 (en) 2006-06-16 2008-01-24 Bittorrent Inc Classification and verification of static file transfer protocols
US7636800B2 (en) 2006-06-27 2009-12-22 International Business Machines Corporation Method and system for memory address translation and pinning
US8244971B2 (en) 2006-07-31 2012-08-14 Google Inc. Memory circuit system and method
US8943245B2 (en) 2006-09-28 2015-01-27 Virident Systems, Inc. Non-volatile type memory modules for main memory
US20080082750A1 (en) 2006-09-28 2008-04-03 Okin Kenneth A Methods of communicating to, memory modules in a memory channel
US8074022B2 (en) 2006-09-28 2011-12-06 Virident Systems, Inc. Programmable heterogeneous memory controllers for main memory with different memory modules
WO2008051940A3 (en) 2006-10-23 2008-08-14 Virident Systems Inc Methods and apparatus of dual inline memory modules for flash memory
WO2008055269A3 (en) 2006-11-04 2008-06-26 Virident Systems Inc Asymmetric memory migration in hybrid main memory
US20080215996A1 (en) * 2007-02-22 2008-09-04 Chad Farrell Media, Llc Website/Web Client System for Presenting Multi-Dimensional Content
US20080229049A1 (en) 2007-03-16 2008-09-18 Ashwini Kumar Nanda Processor card for blade server and process.
CN106407136A (en) 2007-04-12 2017-02-15 拉姆伯斯公司 Memory system with point-to-point request interconnect
US8874831B2 (en) 2007-06-01 2014-10-28 Netlist, Inc. Flash-DRAM hybrid memory module
US8904098B2 (en) 2007-06-01 2014-12-02 Netlist, Inc. Redundant backup using non-volatile memory
US8301833B1 (en) 2007-06-01 2012-10-30 Netlist, Inc. Non-volatile memory module
US8347005B2 (en) 2007-07-31 2013-01-01 Hewlett-Packard Development Company, L.P. Memory controller with multi-protocol interface
US7840748B2 (en) 2007-08-31 2010-11-23 International Business Machines Corporation Buffered memory module with multiple memory device data interface ports supporting double the memory capacity
US7949683B2 (en) 2007-11-27 2011-05-24 Cavium Networks, Inc. Method and apparatus for traversing a compressed deterministic finite automata (DFA) graph
US8438328B2 (en) 2008-02-21 2013-05-07 Google Inc. Emulation of abstracted DIMMs using abstracted DRAMs
US20110235260A1 (en) 2008-04-09 2011-09-29 Apacer Technology Inc. Dram module with solid state disk
US8001434B1 (en) 2008-04-14 2011-08-16 Netlist, Inc. Memory board with self-testing capability
US8154901B1 (en) 2008-04-14 2012-04-10 Netlist, Inc. Circuit providing load isolation and noise reduction
CA2719841C (en) 2008-05-22 2016-03-22 Nokia Siemens Networks Oy Adaptive scheduler for communication systems apparatus, system and method
US8190699B2 (en) 2008-07-28 2012-05-29 Crossfield Technology LLC System and method of multi-path data communications
US7886103B2 (en) 2008-09-08 2011-02-08 Cisco Technology, Inc. Input-output module, processing platform and method for extending a memory interface for input-output operations
JP5272265B2 (en) * 2008-09-29 2013-08-28 株式会社日立製作所 Pci device sharing method
US8054832B1 (en) 2008-12-30 2011-11-08 Juniper Networks, Inc. Methods and apparatus for routing between virtual resources based on a routing location policy
US20100183033A1 (en) 2009-01-20 2010-07-22 Nokia Corporation Method and apparatus for encapsulation of scalable media
US8264903B1 (en) 2009-05-05 2012-09-11 Netlist, Inc. Systems and methods for refreshing a memory module
US8489837B1 (en) 2009-06-12 2013-07-16 Netlist, Inc. Systems and methods for handshaking with a memory module
US8516185B2 (en) 2009-07-16 2013-08-20 Netlist, Inc. System and method utilizing distributed byte-wise buffers on a memory module
US8417870B2 (en) 2009-07-16 2013-04-09 Netlist, Inc. System and method of increasing addressable memory space on a memory board
US9535849B2 (en) 2009-07-24 2017-01-03 Advanced Micro Devices, Inc. IOMMU using two-level address translation for I/O and computation offload devices on a peripheral interconnect
US8848513B2 (en) 2009-09-02 2014-09-30 Qualcomm Incorporated Seamless overlay connectivity using multi-homed overlay neighborhoods
US9389895B2 (en) 2009-12-17 2016-07-12 Microsoft Technology Licensing, Llc Virtual storage target offload techniques
WO2011087820A8 (en) 2009-12-21 2012-03-22 Sanmina-Sci Corporation Method and apparatus for supporting storage modules in standard memory and/or hybrid memory bus architectures
EP2363812B1 (en) 2010-03-04 2018-02-28 Karlsruher Institut für Technologie Reconfigurable processor architecture
KR20130081213A (en) 2010-03-26 2013-07-16 버추얼메트릭스, 인크. Fine grain performance resource management of computer systems
CN101794271B (en) 2010-03-31 2012-05-23 华为技术有限公司 Implementation method and device of consistency of multi-core internal memory
CA2800809A1 (en) 2010-05-28 2011-12-01 Lawrence A. Laurich Accelerator system for use with secure data storage
US8631271B2 (en) 2010-06-24 2014-01-14 International Business Machines Corporation Heterogeneous recovery in a redundant memory system
US9118591B2 (en) 2010-07-30 2015-08-25 Broadcom Corporation Distributed switch domain of heterogeneous components
JP5872560B2 (en) 2010-09-24 2016-03-01 テキサス・メモリー・システムズ・インコーポレイテツド High-speed memory system
WO2012061633A3 (en) 2010-11-03 2013-03-21 Netlist, Inc. Method and apparatus for optimizing driver load in a memory package
US20120239874A1 (en) 2011-03-02 2012-09-20 Netlist, Inc. Method and system for resolving interoperability of multiple types of dual in-line memory modules
US8774213B2 (en) * 2011-03-30 2014-07-08 Amazon Technologies, Inc. Frameworks and interfaces for offload device-based packet processing
US8930647B1 (en) 2011-04-06 2015-01-06 P4tents1, LLC Multiple class memory systems
WO2012141694A1 (en) 2011-04-13 2012-10-18 Hewlett-Packard Development Company, L.P. Input/output processing
US8442056B2 (en) 2011-06-28 2013-05-14 Marvell International Ltd. Scheduling packets in a packet-processing pipeline
US20130019057A1 (en) 2011-07-15 2013-01-17 Violin Memory, Inc. Flash disk array and controller
US8767463B2 (en) 2011-08-11 2014-07-01 Smart Modular Technologies, Inc. Non-volatile dynamic random access memory system with non-delay-lock-loop mechanism and method of operation thereof
US9424188B2 (en) 2011-11-23 2016-08-23 Smart Modular Technologies, Inc. Non-volatile memory packaging system with caching and method of operation thereof
US8918634B2 (en) * 2012-02-21 2014-12-23 International Business Machines Corporation Network node with network-attached stateless security offload device employing out-of-band processing
US9513845B2 (en) 2012-03-30 2016-12-06 Violin Memory Inc. Memory module virtualization
US10019371B2 (en) * 2012-04-27 2018-07-10 Hewlett Packard Enterprise Development Lp Data caching using local and remote memory
US20130318119A1 (en) * 2012-05-22 2013-11-28 Xocketts IP, LLC Processing structured and unstructured data using offload processors
US9378161B1 (en) * 2013-01-17 2016-06-28 Xockets, Inc. Full bandwidth packet handling with server systems including offload processors
US20140281661A1 (en) 2013-03-15 2014-09-18 Netlist, Inc. Hybrid Memory System With Configurable Error Thresholds And Failure Analysis Capability

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20080259555A1 (en) * 2006-01-13 2008-10-23 Sun Microsystems, Inc. Modular blade server
US20070294433A1 (en) * 2006-06-14 2007-12-20 Leigh Kevin B system of implementing switch devices in a server system
US20120047126A1 (en) * 2007-08-27 2012-02-23 Teradata Us, Inc. Methods and systems for hardware acceleration of streamed database operations and queries based on multiple hardware accelerators
US20100031253A1 (en) * 2008-07-29 2010-02-04 Electronic Data Systems Corporation System and method for a virtualization infrastructure management environment
US20100235484A1 (en) * 2009-03-12 2010-09-16 International Business Machines Corporation Remotely Administering A Server
US20110035540A1 (en) * 2009-08-10 2011-02-10 Adtron, Inc. Flash blade system architecture and method
US20120151292A1 (en) * 2010-12-14 2012-06-14 Microsoft Corporation Supporting Distributed Key-Based Processes

Cited By (19)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US9465771B2 (en) 2009-09-24 2016-10-11 Iii Holdings 2, Llc Server on a chip and node cards comprising one or more of same
US9866477B2 (en) 2009-10-30 2018-01-09 Iii Holdings 2, Llc System and method for high-performance, low-power data center interconnect fabric
US9311269B2 (en) 2009-10-30 2016-04-12 Iii Holdings 2, Llc Network proxy for high-performance, low-power data center interconnect fabric
US9405584B2 (en) 2009-10-30 2016-08-02 Iii Holdings 2, Llc System and method for high-performance, low-power data center interconnect fabric with addressing and unicast routing
US9454403B2 (en) 2009-10-30 2016-09-27 Iii Holdings 2, Llc System and method for high-performance, low-power data center interconnect fabric
US9479463B2 (en) 2009-10-30 2016-10-25 Iii Holdings 2, Llc System and method for data center security enhancements leveraging managed server SOCs
US9509552B2 (en) 2009-10-30 2016-11-29 Iii Holdings 2, Llc System and method for data center security enhancements leveraging server SOCs or server fabrics
US9977763B2 (en) 2009-10-30 2018-05-22 Iii Holdings 2, Llc Network proxy for high-performance, low-power data center interconnect fabric
US9929976B2 (en) 2009-10-30 2018-03-27 Iii Holdings 2, Llc System and method for data center security enhancements leveraging managed server SOCs
US9680770B2 (en) 2009-10-30 2017-06-13 Iii Holdings 2, Llc System and method for using a multi-protocol fabric module across a distributed server interconnect fabric
US9749326B2 (en) 2009-10-30 2017-08-29 Iii Holdings 2, Llc System and method for data center security enhancements leveraging server SOCs or server fabrics
US9876735B2 (en) 2009-10-30 2018-01-23 Iii Holdings 2, Llc Performance and power optimized computer system architectures and methods leveraging power optimized tree fabric interconnect
US9262225B2 (en) 2009-10-30 2016-02-16 Iii Holdings 2, Llc Remote memory access functionality in a cluster of data processing nodes
US9585281B2 (en) 2011-10-28 2017-02-28 Iii Holdings 2, Llc System and method for flexible storage and networking provisioning in large scalable processor installations
US10021806B2 (en) 2011-10-28 2018-07-10 Iii Holdings 2, Llc System and method for flexible storage and networking provisioning in large scalable processor installations
US9792249B2 (en) 2011-10-31 2017-10-17 Iii Holdings 2, Llc Node card utilizing a same connector to communicate pluralities of signals
US9965442B2 (en) 2011-10-31 2018-05-08 Iii Holdings 2, Llc Node card management in a modular and large scalable server system
US9648102B1 (en) * 2012-12-27 2017-05-09 Iii Holdings 2, Llc Memcached server functionality in a cluster of data processing nodes
US9934395B2 (en) 2015-09-11 2018-04-03 International Business Machines Corporation Enabling secure big data analytics in the cloud

Also Published As

Publication number Publication date Type
US20170237672A1 (en) 2017-08-17 application
US20130318277A1 (en) 2013-11-28 application
US20130318280A1 (en) 2013-11-28 application
US20140157397A1 (en) 2014-06-05 application
US20130318268A1 (en) 2013-11-28 application
US9558351B2 (en) 2017-01-31 grant
US20170235699A1 (en) 2017-08-17 application
US9286472B2 (en) 2016-03-15 grant
US20140165196A1 (en) 2014-06-12 application

Similar Documents

Publication Publication Date Title
Fusco et al. High speed network traffic analysis with commodity multi-core systems
Ausavarungnirun et al. Staged memory scheduling: Achieving high performance and scalability in heterogeneous systems
US7743191B1 (en) On-chip shared memory based device architecture
US8738860B1 (en) Computing in parallel processing environments
US7934035B2 (en) Apparatus, method and system for aggregating computing resources
Jeong et al. mTCP: a Highly Scalable User-level TCP Stack for Multicore Systems.
US20050240745A1 (en) High speed memory control and I/O processor system
US20090150894A1 (en) Nonvolatile memory (NVM) based solid-state disk (SSD) system for scaling and quality of service (QoS) by parallelizing command execution
US20050081201A1 (en) System and method for grouping processors
US7552298B2 (en) Method and system for deferred pinning of host memory for stateful network interfaces
US7664823B1 (en) Partitioned packet processing in a multiprocessor environment
US20060136681A1 (en) Method and apparatus to support multiple memory banks with a memory block
US20050081181A1 (en) System and method for dynamically partitioning processing across plurality of heterogeneous processors
US7567567B2 (en) Network system including packet classification for partitioned resources
US7941591B2 (en) Flash DIMM in a standalone cache appliance system and methodology
US20060020769A1 (en) Allocating resources to partitions in a partitionable computer
US20060069878A1 (en) System and method for virtualization of processor resources
US7353360B1 (en) Method for maximizing page locality
Franke et al. Introduction to the wire-speed processor and architecture
US20070162619A1 (en) Method and System for Zero Copy in a Virtualized Network Environment
US20090055831A1 (en) Allocating Network Adapter Resources Among Logical Partitions
US7415034B2 (en) Virtualized partitionable shared network interface
Peter et al. Arrakis: The operating system is the control plane
US20120066337A1 (en) Tiered storage interface
US20060221990A1 (en) Hiding system latencies in a throughput networking system

Legal Events

Date Code Title Description
AS Assignment

Owner name: XOCKETS IP, LLC, DELAWARE

Free format text: NUNC PRO TUNC ASSIGNMENT;ASSIGNORS:DALAL, PARIN BHADRIK;BELAIR, STEPHEN PAUL;REEL/FRAME:036663/0441

Effective date: 20150904

Owner name: XOCKETS, INC., CALIFORNIA

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:XOCKETS IP, LLC;REEL/FRAME:036663/0451

Effective date: 20150719