US20090125666A1 - Dynamically scalable queues for performance driven pci express memory traffic - Google Patents

Dynamically scalable queues for performance driven pci express memory traffic Download PDF

Info

Publication number
US20090125666A1
US20090125666A1 US12357098 US35709809A US2009125666A1 US 20090125666 A1 US20090125666 A1 US 20090125666A1 US 12357098 US12357098 US 12357098 US 35709809 A US35709809 A US 35709809A US 2009125666 A1 US2009125666 A1 US 2009125666A1
Authority
US
Grant status
Application
Patent type
Prior art keywords
pci
data
queue
express
entries
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
US12357098
Inventor
Ronald E. Freking
Philip R. Hillier, III
Curtis C. Wollbrink
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.)
International Business Machines Corp
Original Assignee
International Business Machines Corp
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

    • 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
    • G06F13/382Information transfer, e.g. on bus using universal interface adapter
    • G06F13/385Information transfer, e.g. on bus using universal interface adapter for adaptation of a particular data processing system to different peripheral devices

Abstract

A computer program product for implementing a method within a data processing system and a PCI Express protocol for enabling high performance IO data transfers for multiple, different IO configurations, which include variable packet sizes and/or variable/different numbers of transactions on the IO link. PCI Express protocol is enhanced to support utilization of counters and dynamically variable queue sizes. In addition to the standard queue entries, several (or a selected number of) dynamically changeable queue entries are provided/reserved and a dynamic queue modification (DQM) utility is provided within the enhanced PCI Express protocol to monitor ongoing, current data transfer and manage when the size(s) of the queue entries are modified (increased or decreased) based on current data traffic transmitting on the PCI Express IO link. The enhanced PCI Express protocol provides an equilibrium point at which many large data packets are transferred efficiently, while imposing a limit on the number of each size of packets outstanding.

Description

    PRIORITY CLAIM
  • [0001]
    The present application is a continuation of U.S. patent application Ser. No. 11/466,142, filed Aug. 22, 2006 and titled Dynamically Scalable Queues For Performance Driven PCI Express Memory Traffic, the contents of which is incorporated herein by reference.
  • BACKGROUND OF THE INVENTION
  • [0002]
    1. Technical Field
  • [0003]
    The present invention relates generally to computer systems and in particular to input/output operations of computer systems. Still more particularly, the present invention relates to an enhanced method and system for transferring input/output (IO) data from an IO device of a computer system.
  • [0004]
    2. Description of the Related Art
  • [0005]
    Connections of Input/Output (IO) devices to computer systems and the associated transfer of data to and from the connected computer system are typically supported by one of several available hardware devices and associated protocols. In most conventional computer systems, the transfer protocol utilized for local interconnection of these IO devices is Peripheral Component Interconnect (PCI) Express. PCI Express (supported by specific hardware) is an implementation of the PCI computer bus that enables faster physical layer communications via use of a network of serial interconnects (in lieu of a single bus). PCI Express utilizes a single hub with many pins on the mainboard to enable switching and parallel data transfers.
  • [0006]
    The higher speeds accomplished by PCI Express has enabled PCI express to become the new backplane standard in a majority of personal computers. This is also due in part to PCI Express' design, which enables PCI Express to be completely transparent to software developers. Thus, an operating system designed for PCI is able to boot in a PCI Express system without any code modification.
  • [0007]
    Conventional methods for enabling IO data transfer include the utilization of IO queues. However, developing queues for IO devices is currently application specific, particularly when transferring data via PCI Express. PCI Express utilizes a variable size packet-driven serial protocol to transfer data. A queuing structure is required to execute these transfers in a coherent manner. These IO queues are statically configured and support only a single type of data transfer well. For example, if the IO devices that drive the traffic are varied in (1) the sizes of transfers and/or (2) the number of outstanding transactions on the link, developing the queues becomes a choice of exclusively supporting (a) many large transfers, (b) a few large transfers, (c) many small transfers, or (d) a few small transfers. Each category of data transfer operates best at a particular (single) type of queue configuration and losses operational quality for all other types of transfers and corresponding queue configurations.
  • [0008]
    The determination of which queue configuration works best for the particular IO transfer depends on what the computer system (or executing application) requests/requires. Thus, when the system/application is concurrently or sequentially providing different combinations of sizes and numbers of transactions, the statically-configured IO queues are unable to deliver high performance on all of the various configurations. The present invention recognizes and corrects this limitation in the existing IO data transfer methods, particularly those that utilize PCI Express.
  • SUMMARY OF THE INVENTION
  • [0009]
    Disclosed is a computer program product that implements a method on a computer system and a PCI Express device/protocol for enabling high performance IO data transfers for multiple, different IO configurations, which include variable packet sizes and/or variable/different numbers of transactions on the IO link. PCI Express protocol is enhanced to support utilization of counters and dynamically variable queue sizes. In addition to the standard queue entries, several (or a selected number of) dynamically changeable queue entries are provided/reserved and a dynamic queue modification (DQM) logic is provided within the enhanced PCI Express protocol. The DQM logic monitors ongoing, current data transfer and manages when the size(s) of the queue entries are modified (increased or decreased) based on current data traffic transmitting on the PCI Express IO link.
  • [0010]
    When data traffic tends towards a single stream of large data packets, queue entries are automatically combined and utilized to transfer the large data as quickly as possible. However, if the data traffic tends towards smaller data packets, the queue entries are broken up into many independent entries to handle the individual, smaller data packets. The enhanced PCI Express protocol provides an equilibrium point at which many large data packets are transferred efficiently, while imposing a limit on the number of each size of packets outstanding.
  • [0011]
    The above as well as additional objectives, features, and advantages of the present invention will become apparent in the following detailed written description.
  • BRIEF DESCRIPTION OF THE DRAWINGS
  • [0012]
    The invention itself, as well as a preferred mode of use, further objects, and advantages thereof, will best be understood by reference to the following detailed description of an illustrative embodiment when read in conjunction with the accompanying drawings, wherein:
  • [0013]
    FIG. 1 is a block diagram of a computer system having an enhanced PCI Express devices and protocol for handling variable sized IO traffic utilizing dynamically configurable IO queues according to one embodiment of the invention;
  • [0014]
    FIG. 2 is a more detailed illustration of a PCI Express subsystem with dynamic queue modification (DQM) logic and dynamically-changeable queues, according to one embodiment of the invention;
  • [0015]
    FIGS. 3A-3B are flow charts illustrating the processes by which the enhanced PCI Express device and protocol re-configures the IO queues (or queues entries) based on current traffic flow, according to one embodiment of the invention; and
  • [0016]
    FIG. 4 illustrates an example sequence of dynamically configurable IO queue entries with a maximum queue size and variable numbers of queue entries, according to one embodiment of the invention.
  • DETAILED DESCRIPTION OF AN ILLUSTRATIVE EMBODIMENT
  • [0017]
    The present invention provides a computer program product that implements a method on a computer system and a PCI Express device/protocol for enabling high performance IO data transfers for multiple, different IO configurations, which include variable packet sizes and/or variable/different numbers of transactions on the IO link. PCI Express protocol is enhanced to support utilization of counters and dynamically variable queue sizes. In addition to the standard queue entries, several (or a selected number of) dynamically changeable queue entries are provided/reserved and a dynamic queue modification (DQM) logic is provided within the enhanced PCI Express protocol. The DQM logic monitors ongoing, current data transfer and manages when the size(s) of the queue entries are modified (increased or decreased) based on current data traffic transmitting on the PCI Express IO link.
  • [0018]
    When data traffic tends towards a single stream of large data packets, queue entries are automatically combined and utilized to transfer the large data as quickly as possible. However, if the data traffic tends towards smaller data packets, the queue entries are broken up into many independent entries to handle the individual, smaller data packets. The enhanced PCI Express protocol provides an equilibrium point at which many large data packets are transferred efficiently, while imposing a limit on the number of each size of packets outstanding.
  • [0019]
    With reference now to the figures, and in particular to FIG. 1, which is a block diagram representation of a data processing system configured with enhanced PCI Express device and associated protocol for completing the dynamic (on-the-fly) size modification of IO queue entries based on current (or existing or received) IO traffic, in accordance with the illustrative embodiment of the present invention. Data processing system 100 comprises at least one central processing unit (CPU) 102 (or processor) connected to a system bus 104. In alternate embodiments, data processing system 100 may be a multiprocessor (MP) system with a plurality of processors. Also connected to system bus 104 is memory controller 106, which provides an interface to system memory 108. I/O bus controller 110 is connected to system bus 104 and provides an interface to I/O bus 112. Peripheral component interconnect (PCI) Express bus controller 114 is connected to I/O bus 112 and provides an interface to PCI Express local bus 116. PCI Express bus implementation may support four PCI expansion slots or add-in connectors. A number of modems/adapters may be connected to PCI Express local bus 116, enabling communication to IO devices. Existing PCI Express firmware is enhanced with DQM logic 118, which enables the various features of the invention described below.
  • [0020]
    As provided within the invention, the PCI Express specification/protocol defines data sizes to be between 1 to 4096 Bytes in length. The transfer of this range of data sizes in a coherent structure requires large packets be broken into manageable chunks. With conventional implementation, if a queue has many entries for small data packets, a large data packet would consume all the queue entries or take a long time to execute when/if one queue entry was utilized to transfer the entire data packet. However, with the invention, the sizes of these queues are variable and depend on the type of traffic actually received (rather then statically configured for the traffic that is expected). Also, the queues are broken into entries that are able to handle the data expected. These entries are either many small entries or a few large entries or a combination of both.
  • [0021]
    In addition to the above hardware components, various features of the invention may be implemented via software executing on CPU 102. For illustrative purposes, those software components are represented within system memory 109 as operating system (OS) 120 and applications 122. Program code of OS 120 and applications 122 execute on CPU 102 and may generate IO data that is transferred to a PCI connected device (not specifically shown) via PCI bus controller 114. Actual processing/manipulation of data during data transfer from the processor to the PCI device via/through the PCI fabric, as well as processing/manipulation of data received from the connected PCI device utilizing dynamically configurable IO queues is described in detail below.
  • [0022]
    FIG. 2 illustrates an IO view of PCI Express subsystem 200 with attached PCI devices, according to one embodiment. PCI express subsystem 200 may be implementing within data processing system 100, and similar components are provided the same reference numerals between the figures. Within this illustration, CPU 102 is coupled to PCI Express switch 206 via first buses 204, and PCI bridge 212 is connected to PCI Express switch 206 via second buses 210. PCI Express switch 206 may be also coupled via third buses 208 to other PCI Express devices or switches (not shown). In the illustrative embodiment, PCI Express switch 206 comprises DQM logic 118, which is utilized to dynamically size PCI Express queues 216. PCI Express switch 206 and PCI bridge 212 may collectively make up PCI Express bus controller 114 of FIG. 1. In alternate embodiments, first buses 204 may also provide DQM functions when first buses are PCI Express buses. Also, as provided in the illustrative embodiment, PCI Express interface 214 also comprises DQM logic 118. The specific location of DQM logic within the general PCI Express bus controller 114 may vary based on system design and it is contemplated that either a single logic or a distributed logic may be utilized within different embodiments of the invention.
  • [0023]
    PCI bridge 212 is also coupled to PCI bus 116, to which several PCI compliant devices 232, 234, 236 are coupled. Each of these PCI compliant devices 232, 234, 236 has a request/grant pair of buses 230, 228, 226, respectively. The request/grant buses 230, 228, 226 are coupled to PCI bus arbiter 220. PCI bus 116 is coupled to PCI bus interface 218. PCI bus interface 218 is coupled by a two way bus to PCI transaction queues 216. PCI transaction queues 216 are shown in greater details in FIG. 4, which is described below. PCI bus interface 218 is also coupled by request/grant pair of busses 222 to PCI bus arbiter 220. Additional information supplied by PCI bus arbiter 220 and which may be stored with the data in PCI transaction queues 216 is provided to PCI transaction queues 216 along arbiter bus 238. PCI transaction queues 216 are coupled by a two way bus to the PCI Express interface 214, which is coupled to PCI Express switch 206 via second buses 210. PCI Express interface 214 and/or PCI Express switch 206 provide queue entry modification commands from DQM logic 118 to PCI transaction queues 216. While the illustrative embodiment is shown with PCI devices connected to PCI Express, the functional features of the invention are applicable to other types of configurations. Those skilled in the art will appreciate that the illustrative embodiment I provided solely for illustration and not intended to imply any limitations to the applicability of the invention.
  • [0024]
    In operation, when the slave circuit in PCI bus interface 218 receives the data, the circuit processes the data into the PCI transaction queues. Concurrently, DQM logic 118 performs the necessary operations to modify queue sizes, if required, based on the data characteristics. The process by which the size(s) of the queue entries are modified is described in details below with reference to FIG. 3.
  • [0025]
    Those of ordinary skill in the art will appreciate that the hardware depicted in FIGS. 1 and 2 may vary. For example, other peripheral devices, such as optical disk drives and the like, also may be used in addition to or in place of the hardware depicted. Thus, the depicted example is not meant to imply architectural limitations with respect to the present invention. The data processing system depicted in FIGS. 1 and 2 may be, for example, an IBM eServer pSeries system, a product of International Business Machines Corporation in Armonk, N.Y., running the Advanced Interactive Executive (AIX) operating system or LINUX operating system.
  • [0026]
    FIGS. 3A-3B provide flow charts illustrating the operations performed/undertaken during processing of data that is being transferred to and from a PCI (or other) device via PCI Express bus (controller) 114 with associated PCI transaction queues 216. According to the invention, the queues comprise one or more entries that are dynamically configurable, based on the type and number of data being transferred through the PCI fabric.
  • [0027]
    The number of transactions supported and queue entries provided are finite based on a maximum amount of available space allocated to the queues. In the exemplary embodiment, the maximum size single queue supported contains a single queue entry of 2096 bytes, and the queue is capable of supporting up to 32 entries at 64 bytes each. Also, in the described embodiment, the queue entries are sized at 512 bytes each, resulting in only four (4) queue entries being available during system initialization (device power on or setup).
  • [0028]
    Processing of the inventive steps is completed via DQM logic 118 once the initial parameters are established. According to the invention, DQM logic 118 comprises one or more performance counters, which take on one of several predefined characteristics based on the number of outstanding transactions (NTR). To understand the processing completed by the utility, the following parameters are defined and utilized within the example DQM logic 118:
      • NTR Number of transactions
      • N64 Number of transactions between 0 B and 64 B
      • N256 Number of transactions between 64 B and 256 B
      • N512 Number of transactions between 256 B and 512 B
      • Nxxx Number of transactions (up to Max Payload)
      • RLT Reset Limit
      • RLEn Reset Once
      • P(1-6) % Percent used in “If” statements (1-6), where six different threshold percentages are established and utilized to determine when to modify a queue entry
  • [0037]
    Parameter NTR is reset after reaching the reset limit (RLT). In the specific illustrative embodiment, when NTR reaches RLT, an evaluation of queue entry sizes is completed by DQM logic 118. This evaluation may occur only once if RLEn is set to “on” (e.g., the value of RLEn=1). However, when RLEn is not set to “on,” (e.g., the value of RLEn=0), the evaluation of queue entry sizes is triggered and completed every time NTR reaches RLT. DQM logic 118 performs the consolidation or expansion of queue entries within queues 216 based on a series of conditions, which are presented below. Each condition includes a determination of the number of outstanding transactions, where P(x) % is the maximum threshold percentage for that particular number of transactions to trigger the resize operation. Thus:
      • (a) If N64 is greater than P1% of NTR, set the queues to 64 B entries;
      • (b) If N256 is greater than P2% of NTR, set the queues to 256 B entries;
      • (c) If N512 is greater than P3% of NTR, set the queues to 512 B entries;
      • (d) If N64+N512 is greater than P4% of NTR set the queues to 256 B;
      • (e) If N256+N512 is greater than P5% or NTR set queues to 256 B;
      • (f) If N256+N64 is greater than P6% of NTR set queues to 128 B; and
      • (g) If none of the above proves to be true, the utility leaves the queues in the current configuration.
  • [0045]
    FIG. 4 illustrates example queue entry sizes within a single queue 410. In the illustration, the queue entry sizes are based on the maximum queue size (e.g., 2096 B) and the number of outstanding data of a particular size (requiring a specific size entry). Only those queues that are designated as changeable are affected by DQM logic 118, and the number of queues designated as changeable may be a design parameter or a programmable parameter based on the executing application(s). Thus, some of the available queues may contain entries with preset sizes that are not changed by the processes of DQM logic 118. Returning to the figure, each entry within the set of queue entries (from the leftmost queue 410 proceeding sequentially to the rightmost queue 410) is divided down by a factor of two as the size of the data (data packets) decreases. Also, the illustrative embodiment indicates that the entire queue 410 is not necessarily divided into equal parts, thus allowing for both smaller packets and larger packets to be queued within a single queue 410.
  • [0046]
    Turning now to FIG. 3A, the process begins at block 302 at which the queues (and queue entries) are initially configured with a default queue size based on the minimum data transfer and the maximum outstanding transactions that may be handled by the queue. These parameters may be preset or determined during activation of the PCI Express IO devices and/or applications accessing the IO devices. DQM logic 118 monitors for commands to begin flowing across the PCI express link, as shown at block 304, and as commands begin flowing across the PCI express link, DQM logic 118 activates a series of counters (or counter mechanism) to track specific counter parameters, as indicated at block 306. According to the illustrative embodiment, the counter is utilized to keep track of (a) the number of transactions outstanding at one time, (b) the average size of transactions, and (c) the number of queue entries available (hereinafter collectively referred to as the “counter tracking parameters”).
  • [0047]
    At block 308, the counter values are compared to a set of pre-established threshold values (one for each counter), and a determination is made at block 310 whether the counter values have reach the associated threshold. If any one of the counter values have reached the associated threshold, DQM logic 118 initiates a resizing of the queue entries to accommodate for that particular sized data packet, as shown by block 312.
  • [0048]
    FIG. 3B depicts a more detailed evaluation performed by DQM logic 118 in determining whether to reduce or enlarge the size of the queue entries. This determination begins at block 314, at which DQM logic 118 checks whether there are a greater number of smaller data packets outstanding than the associated threshold for that size packet. When there is a greater number of outstanding smaller data packets than the associated threshold, DQM logic 118 splits the queue entries, as shown at block 316, to generate smaller queue entries to accommodate a larger number of smaller data packets. For example, if there are four data packets outstanding and the sizes of the packets are 256 bytes or less, DQM logic 118 splits the initial four queue entries (originally 512 B each) into 8 entries at 256 bytes, as shown. DQM logic 118 thus divides down the queue to a maximum number of entries that has been defined. In one embodiment, a maximum number of entries may be established to drive the determination.
  • [0049]
    Returning to decision block 314, if DQM logic 118 determines that the number of smaller data packets outstanding is not greater than the associated threshold for that size packet, DQM logic 118 checks, as depicted at block 318, whether there are a greater number of larger data packets outstanding than the associated threshold for that size packet. When there is a greater number of outstanding larger data packets than the associated threshold, DQM logic 118 combines or consolidates two or more or the queue entries, as shown at block 320, to generate larger queue entries to accommodate a greater number of larger data packets. For example, assuming there are four queue entries having an initial 512 byte capacity and there are only two large data transfers outstanding at a time, DQM logic 118 consolidates the queue to two (2) entries at 1024 bytes each. The performance counters are thus utilized to show trends in the data traffic, and DQM logic 118 is eventually able to change the queue sizes based on the performance counters in order to gain better performance in different IO data transfer situations. The monitoring and resizing of queue entries continues for additional data traffic received as indicated by the return to block 304 from each end point of the sub-process.
  • [0050]
    As a final matter, it is important that while an illustrative embodiment of the present invention has been, and will continue to be, described in the context of a fully functional computer system with installed management software, those skilled in the art will appreciate that the software aspects of an illustrative embodiment of the present invention are capable of being distributed as a program product in a variety of forms, and that an illustrative embodiment of the present invention applies equally regardless of the particular type of signal bearing media used to actually carry out the distribution. Examples of signal bearing media include recordable type media such as floppy disks, thumb drives, hard disk drives, CD ROMs, DVDs, and transmission type media such as digital and analogue communication links.
  • [0051]
    While the invention has been particularly shown and described with reference to a preferred embodiment, it will be understood by those skilled in the art that various changes in form and detail may be made therein without departing from the spirit and scope of the invention.

Claims (9)

  1. 1. A computer program product comprising:
    a computer readable storage medium; and
    program code stored on the computer readable storage medium that when executed by a processor of a computer device in which the processor is coupled to a system bus and a peripheral component interconnect (PCI) fabric, the program code enables the computer device to perform functions for managing transfer of data packets across the PCI fabric, said functions including:
    monitoring ongoing, current data transfer within the PCI fabric, wherein the PCI fabric includes a PCI Express bus controller coupled to the system bus, said PCI express bus controller having one or more PCI queues with one or more queue entries and a dynamic queue modification (DQM) logic; and
    triggering a dynamic modification of a size of each queue entry based on current data traffic transmitting on the PCI fabric, wherein said dynamic modification is completed via a DQM logic.
  2. 2. The computer program product of claim 1, wherein:
    said monitoring comprises code for determining a current trend within ongoing data traffic, said trend ranging from a single stream of a large data packet to multiple streams of smaller data packets; and
    said triggering comprises:
    when the trend is that of a single stream of large data packets, automatically combining smaller queue entries into larger entries to support efficient transfer of the larger data packets across the PCI fabric; and
    when the trend is that of multiple streams of smaller data packets, automatically breaking larger queue entries into smaller independent entries to support efficient transfer of the individual, smaller data packets.
  3. 3. The computer program product of claim 1, wherein said DQM logic includes one or more performance counters, which take on one of several predefined characteristics based on the number of outstanding transactions (NTR), wherein said counters are utilized to track a number of received transactions in a pre-specified ranges of transaction size, said program code further comprising code for performing the functions of:
    resetting a counter when the counter reaches a preset limit;
    tracking a percentage of transactions within each of said pre-specified ranges of transaction sizes that is currently outstanding and waiting to be transferred within the PCI fabric, wherein when a percentage of existing transactions passes a preset threshold, said DQM logic triggers said dynamic configuration of the queue entries to support a transfer of more transactions within that range of specific transaction size; and
    consolidating and expanding queue entries within the queue based on the tracked percentage of transactions within each of the pre-specified ranges of transaction sizes.
  4. 4. The computer program product of claim 1, further comprising program code for performing the functions of:
    initially configuring each queue entry within the queue to a default queue size, said default queue size based on a minimum data transfer and a maximum number of outstanding transactions that may be handled by the queue;
    monitoring for transactions flowing across the PCI express fabric;
    activating and updating a series of counters to track specific counter parameters from among (a) the number of transactions outstanding at one time, (b) the average size of transactions, and (c) the number of queue entries available;
    comparing counter values to a set of pre-established threshold values; and
    when one of the counter values has reach the corresponding, pre-established threshold, initiating a resizing of the queue entries to accommodate a larger number of transaction packets corresponding to the threshold value.
  5. 5. The computer program product of claim 1, wherein the DQM utility further comprises:
    logic for enabling input/output (IO) data transfers with multiple, different IO data packet characteristics and configurations from among (a) variable, different sizes of data packets and (b) variable numbers of data packets of each size waiting to be schedule for transfer through the PCI fabric; and
    a plurality of counters, including a first set of counters, which each track a number of queue entries of a particular size and a second set of counters which each track a number of outstanding data packets of each different size supported by the PCI fabric.
  6. 6. The computer program product of claim 1, wherein:
    the PCI Express bus controller of the data processing system comprises multiple queues with at least a first static queue having a static number of queue entries of a particular size and at least a second dynamically configurable queue with a variable number of dynamically changeable queue entries that are modified in real time by the DQM logic; and
    said program code comprises code for: (a) monitoring ongoing, current data transfer; and (b) triggering a modification of a size of each queue entry based on current number of outstanding data traffic waiting to be transferred over the PCI fabric.
  7. 7. The computer program product of claim 1, further comprising program code for completing the functions of:
    establishing thresholds associated with each size of data packets for triggering when to modify the size of the queue entries to that corresponding to the particular size of data packets, wherein said threshold establishes a number above which the number of outstanding data packets of the corresponding size should be reduced;
    comparing a current number of outstanding data packets of each size against the threshold corresponding to that size data packet; and
    adjusting the rate at which a size of a queue entry is modified based on an analysis of the current traffic relative to a pre-established equilibrium point that balances (a) efficient transfer of a number of larger data packets with (b) limiting a number of each size of data packets outstanding to below the preset thresholds of the specific size of data packets.
  8. 8. The computer program product of claim 1, further comprising program code for completing the functions of:
    determining whether there are a greater number of smaller data packets outstanding than the associated threshold for that size packet;
    when there is a greater number of outstanding smaller data packets than the corresponding, pre-established threshold, splitting larger queue entries to generate smaller queue entries that accommodate a larger number of smaller data packets;
    determining whether there are a greater number of larger data packets outstanding than the associated threshold for the larger data packets; and
    when there is a greater number of outstanding larger data packets than the associated threshold, combining two or more or the existing queue entries to generate larger queue entries to accommodate a greater number of larger data packets.
  9. 9. The computer program product of claim 1, wherein said data processing system further comprises:
    an input/output (I/O) bus controller connected to the system bus and providing an I/O bus, wherein said PCI Express bus controller is coupled to the I/O bus and provides an interface to the PCI Express local bus; and
    wherein said PCI Express bus controller further comprises a PCI Express switch, a PCI Express interface, and PCI Express buses, wherein said DQM logic is within one or more of the PCI Express switch, the PCI Express interface, and the PCI Express buses.
US12357098 2006-08-22 2009-01-21 Dynamically scalable queues for performance driven pci express memory traffic Abandoned US20090125666A1 (en)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US11466142 US7496707B2 (en) 2006-08-22 2006-08-22 Dynamically scalable queues for performance driven PCI express memory traffic
US12357098 US20090125666A1 (en) 2006-08-22 2009-01-21 Dynamically scalable queues for performance driven pci express memory traffic

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
US12357098 US20090125666A1 (en) 2006-08-22 2009-01-21 Dynamically scalable queues for performance driven pci express memory traffic

Publications (1)

Publication Number Publication Date
US20090125666A1 true true US20090125666A1 (en) 2009-05-14

Family

ID=39129490

Family Applications (2)

Application Number Title Priority Date Filing Date
US11466142 Expired - Fee Related US7496707B2 (en) 2006-08-22 2006-08-22 Dynamically scalable queues for performance driven PCI express memory traffic
US12357098 Abandoned US20090125666A1 (en) 2006-08-22 2009-01-21 Dynamically scalable queues for performance driven pci express memory traffic

Family Applications Before (1)

Application Number Title Priority Date Filing Date
US11466142 Expired - Fee Related US7496707B2 (en) 2006-08-22 2006-08-22 Dynamically scalable queues for performance driven PCI express memory traffic

Country Status (2)

Country Link
US (2) US7496707B2 (en)
CN (1) CN101132361A (en)

Cited By (31)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011160719A1 (en) 2010-06-23 2011-12-29 International Business Machines Corporation Measurement facility for adapter functions
US8417911B2 (en) 2010-06-23 2013-04-09 International Business Machines Corporation Associating input/output device requests with memory associated with a logical partition
US8416834B2 (en) 2010-06-23 2013-04-09 International Business Machines Corporation Spread spectrum wireless communication code for data center environments
US8447901B2 (en) 2011-02-18 2013-05-21 Ab Initio Technology Llc Managing buffer conditions through sorting
US8458387B2 (en) 2010-06-23 2013-06-04 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification to a guest operating system
US8478922B2 (en) 2010-06-23 2013-07-02 International Business Machines Corporation Controlling a rate at which adapter interruption requests are processed
US8505032B2 (en) 2010-06-23 2013-08-06 International Business Machines Corporation Operating system notification of actions to be taken responsive to adapter events
US8504754B2 (en) 2010-06-23 2013-08-06 International Business Machines Corporation Identification of types of sources of adapter interruptions
US8510599B2 (en) 2010-06-23 2013-08-13 International Business Machines Corporation Managing processing associated with hardware events
US8549182B2 (en) 2010-06-23 2013-10-01 International Business Machines Corporation Store/store block instructions for communicating with adapters
US8566480B2 (en) 2010-06-23 2013-10-22 International Business Machines Corporation Load instruction for communicating with adapters
US8572635B2 (en) 2010-06-23 2013-10-29 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification
US8615645B2 (en) 2010-06-23 2013-12-24 International Business Machines Corporation Controlling the selectively setting of operational parameters for an adapter
US8615622B2 (en) 2010-06-23 2013-12-24 International Business Machines Corporation Non-standard I/O adapters in a standardized I/O architecture
US8621112B2 (en) 2010-06-23 2013-12-31 International Business Machines Corporation Discovery by operating system of information relating to adapter functions accessible to the operating system
US8626970B2 (en) 2010-06-23 2014-01-07 International Business Machines Corporation Controlling access by a configuration to an adapter function
US8631222B2 (en) 2010-06-23 2014-01-14 International Business Machines Corporation Translation of input/output addresses to memory addresses
US8639858B2 (en) 2010-06-23 2014-01-28 International Business Machines Corporation Resizing address spaces concurrent to accessing the address spaces
US8645606B2 (en) 2010-06-23 2014-02-04 International Business Machines Corporation Upbound input/output expansion request and response processing in a PCIe architecture
US8645767B2 (en) 2010-06-23 2014-02-04 International Business Machines Corporation Scalable I/O adapter function level error detection, isolation, and reporting
US8650337B2 (en) 2010-06-23 2014-02-11 International Business Machines Corporation Runtime determination of translation formats for adapter functions
US8656228B2 (en) 2010-06-23 2014-02-18 International Business Machines Corporation Memory error isolation and recovery in a multiprocessor computer system
US8671287B2 (en) 2010-06-23 2014-03-11 International Business Machines Corporation Redundant power supply configuration for a data center
US8677180B2 (en) 2010-06-23 2014-03-18 International Business Machines Corporation Switch failover control in a multiprocessor computer system
US8683108B2 (en) 2010-06-23 2014-03-25 International Business Machines Corporation Connected input/output hub management
US8745292B2 (en) 2010-06-23 2014-06-03 International Business Machines Corporation System and method for routing I/O expansion requests and responses in a PCIE architecture
US8918573B2 (en) 2010-06-23 2014-12-23 International Business Machines Corporation Input/output (I/O) expansion response processing in a peripheral component interconnect express (PCIe) environment
US9003084B2 (en) * 2011-02-18 2015-04-07 Ab Initio Technology Llc Sorting
US9195623B2 (en) 2010-06-23 2015-11-24 International Business Machines Corporation Multiple address spaces per adapter with address translation
US9213661B2 (en) 2010-06-23 2015-12-15 International Business Machines Corporation Enable/disable adapters of a computing environment
US9342352B2 (en) 2010-06-23 2016-05-17 International Business Machines Corporation Guest access to address spaces of adapter

Families Citing this family (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7571287B2 (en) 2003-03-13 2009-08-04 Marvell World Trade Ltd. Multiport memory architecture, devices and systems including the same, and methods of using the same
US8140827B2 (en) * 2007-06-19 2012-03-20 Samsung Electronics Co., Ltd. System and method for efficient data transmission in a multi-processor environment
US8234425B1 (en) 2007-06-27 2012-07-31 Marvell International Ltd. Arbiter module
US7949817B1 (en) 2007-07-31 2011-05-24 Marvell International Ltd. Adaptive bus profiler
US8131915B1 (en) 2008-04-11 2012-03-06 Marvell Intentional Ltd. Modifying or overwriting data stored in flash memory
US8683085B1 (en) 2008-05-06 2014-03-25 Marvell International Ltd. USB interface configurable for host or device mode
US8423710B1 (en) 2009-03-23 2013-04-16 Marvell International Ltd. Sequential writes to flash memory
US8213236B1 (en) 2009-04-21 2012-07-03 Marvell International Ltd. Flash memory
US8688922B1 (en) 2010-03-11 2014-04-01 Marvell International Ltd Hardware-supported memory management
US8756394B1 (en) 2010-07-07 2014-06-17 Marvell International Ltd. Multi-dimension memory timing tuner
CN102340442B (en) * 2010-07-21 2015-07-22 杭州华三通信技术有限公司 Method for adjusting interface receiving queue length and apparatus thereof
US8549504B2 (en) * 2010-09-25 2013-10-01 Intel Corporation Apparatus, method, and system for providing a decision mechanism for conditional commits in an atomic region
US9146776B1 (en) 2011-08-16 2015-09-29 Marvell International Ltd. Systems and methods for controlling flow of message signaled interrupts
US9128920B2 (en) * 2011-11-30 2015-09-08 Marvell World Trade Ltd. Interrupt handling systems and methods for PCIE bridges with multiple buses
DE102012014681B4 (en) * 2012-05-29 2017-01-26 Balluff Gmbh Using a lO links for connecting a power supply unit
CN103502925B (en) * 2012-12-21 2016-05-25 华为技术有限公司 A monitoring method and apparatus for recording management
US9288163B2 (en) * 2013-03-15 2016-03-15 Avago Technologies General Ip (Singapore) Pte. Ltd. Low-latency packet receive method for networking devices
US20150032961A1 (en) * 2013-07-23 2015-01-29 Lexmark International Technologies S.A. System and Methods of Data Migration Between Storage Devices
US9465768B2 (en) 2014-03-14 2016-10-11 International Business Machines Corporation PCI function measurement block enhancements
US9990139B2 (en) * 2015-04-16 2018-06-05 Samsung Electronics Co., Ltd. Scalable and area optimized method to implement command queues in SRIOV based NVM devices

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6044418A (en) * 1997-06-30 2000-03-28 Sun Microsystems, Inc. Method and apparatus for dynamically resizing queues utilizing programmable partition pointers
US6145061A (en) * 1998-01-07 2000-11-07 Tandem Computers Incorporated Method of management of a circular queue for asynchronous access
US20020087758A1 (en) * 2000-11-15 2002-07-04 Dixon Robert William Dynamic buffer size allocation for multiplexed streaming
US20050086448A1 (en) * 2003-10-16 2005-04-21 International Business Machines Corporation System and method of adaptively reconfiguring buffers
US20050240745A1 (en) * 2003-12-18 2005-10-27 Sundar Iyer High speed memory control and I/O processor system
US20060133396A1 (en) * 2004-12-20 2006-06-22 Shah Hemal V Managing system memory resident device management queues
US20060153179A1 (en) * 2004-12-28 2006-07-13 Michael Ho Techniques for transmitting and receiving traffic over advanced switching compatible switch fabrics
US20060174251A1 (en) * 2005-02-03 2006-08-03 Level 5 Networks, Inc. Transmit completion event batching
US20060224920A1 (en) * 2005-03-31 2006-10-05 Intel Corporation (A Delaware Corporation) Advanced switching lost packet and event detection and handling
US7327749B1 (en) * 2004-03-29 2008-02-05 Sun Microsystems, Inc. Combined buffering of infiniband virtual lanes and queue pairs

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7093037B2 (en) * 2003-09-22 2006-08-15 Emulex Design & Manufacturing Corporation Generalized queue and specialized register configuration for coordinating communications between tightly coupled processors

Patent Citations (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6044418A (en) * 1997-06-30 2000-03-28 Sun Microsystems, Inc. Method and apparatus for dynamically resizing queues utilizing programmable partition pointers
US6145061A (en) * 1998-01-07 2000-11-07 Tandem Computers Incorporated Method of management of a circular queue for asynchronous access
US20020087758A1 (en) * 2000-11-15 2002-07-04 Dixon Robert William Dynamic buffer size allocation for multiplexed streaming
US20050086448A1 (en) * 2003-10-16 2005-04-21 International Business Machines Corporation System and method of adaptively reconfiguring buffers
US7143263B2 (en) * 2003-10-16 2006-11-28 International Business Machines Corporation System and method of adaptively reconfiguring buffers
US20050240745A1 (en) * 2003-12-18 2005-10-27 Sundar Iyer High speed memory control and I/O processor system
US7327749B1 (en) * 2004-03-29 2008-02-05 Sun Microsystems, Inc. Combined buffering of infiniband virtual lanes and queue pairs
US20060133396A1 (en) * 2004-12-20 2006-06-22 Shah Hemal V Managing system memory resident device management queues
US20060153179A1 (en) * 2004-12-28 2006-07-13 Michael Ho Techniques for transmitting and receiving traffic over advanced switching compatible switch fabrics
US20060174251A1 (en) * 2005-02-03 2006-08-03 Level 5 Networks, Inc. Transmit completion event batching
US20060224920A1 (en) * 2005-03-31 2006-10-05 Intel Corporation (A Delaware Corporation) Advanced switching lost packet and event detection and handling

Cited By (45)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2011160719A1 (en) 2010-06-23 2011-12-29 International Business Machines Corporation Measurement facility for adapter functions
US8417911B2 (en) 2010-06-23 2013-04-09 International Business Machines Corporation Associating input/output device requests with memory associated with a logical partition
US8416834B2 (en) 2010-06-23 2013-04-09 International Business Machines Corporation Spread spectrum wireless communication code for data center environments
US9383931B2 (en) 2010-06-23 2016-07-05 International Business Machines Corporation Controlling the selectively setting of operational parameters for an adapter
US8457174B2 (en) 2010-06-23 2013-06-04 International Business Machines Corporation Spread spectrum wireless communication code for data center environments
US8458387B2 (en) 2010-06-23 2013-06-04 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification to a guest operating system
US8468284B2 (en) 2010-06-23 2013-06-18 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification to a guest operating system
US8478922B2 (en) 2010-06-23 2013-07-02 International Business Machines Corporation Controlling a rate at which adapter interruption requests are processed
US8505032B2 (en) 2010-06-23 2013-08-06 International Business Machines Corporation Operating system notification of actions to be taken responsive to adapter events
US8504754B2 (en) 2010-06-23 2013-08-06 International Business Machines Corporation Identification of types of sources of adapter interruptions
US8510599B2 (en) 2010-06-23 2013-08-13 International Business Machines Corporation Managing processing associated with hardware events
US8549182B2 (en) 2010-06-23 2013-10-01 International Business Machines Corporation Store/store block instructions for communicating with adapters
US8566480B2 (en) 2010-06-23 2013-10-22 International Business Machines Corporation Load instruction for communicating with adapters
US8572635B2 (en) 2010-06-23 2013-10-29 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification
US8601497B2 (en) 2010-06-23 2013-12-03 International Business Machines Corporation Converting a message signaled interruption into an I/O adapter event notification
US8615645B2 (en) 2010-06-23 2013-12-24 International Business Machines Corporation Controlling the selectively setting of operational parameters for an adapter
US8615622B2 (en) 2010-06-23 2013-12-24 International Business Machines Corporation Non-standard I/O adapters in a standardized I/O architecture
US8621112B2 (en) 2010-06-23 2013-12-31 International Business Machines Corporation Discovery by operating system of information relating to adapter functions accessible to the operating system
US8626970B2 (en) 2010-06-23 2014-01-07 International Business Machines Corporation Controlling access by a configuration to an adapter function
US8631222B2 (en) 2010-06-23 2014-01-14 International Business Machines Corporation Translation of input/output addresses to memory addresses
US8635430B2 (en) 2010-06-23 2014-01-21 International Business Machines Corporation Translation of input/output addresses to memory addresses
US8639858B2 (en) 2010-06-23 2014-01-28 International Business Machines Corporation Resizing address spaces concurrent to accessing the address spaces
US9195623B2 (en) 2010-06-23 2015-11-24 International Business Machines Corporation Multiple address spaces per adapter with address translation
US8645767B2 (en) 2010-06-23 2014-02-04 International Business Machines Corporation Scalable I/O adapter function level error detection, isolation, and reporting
US8650337B2 (en) 2010-06-23 2014-02-11 International Business Machines Corporation Runtime determination of translation formats for adapter functions
US8650335B2 (en) 2010-06-23 2014-02-11 International Business Machines Corporation Measurement facility for adapter functions
US8656228B2 (en) 2010-06-23 2014-02-18 International Business Machines Corporation Memory error isolation and recovery in a multiprocessor computer system
US8671287B2 (en) 2010-06-23 2014-03-11 International Business Machines Corporation Redundant power supply configuration for a data center
US8677180B2 (en) 2010-06-23 2014-03-18 International Business Machines Corporation Switch failover control in a multiprocessor computer system
US8683108B2 (en) 2010-06-23 2014-03-25 International Business Machines Corporation Connected input/output hub management
US8700959B2 (en) 2010-06-23 2014-04-15 International Business Machines Corporation Scalable I/O adapter function level error detection, isolation, and reporting
US8745292B2 (en) 2010-06-23 2014-06-03 International Business Machines Corporation System and method for routing I/O expansion requests and responses in a PCIE architecture
US9342352B2 (en) 2010-06-23 2016-05-17 International Business Machines Corporation Guest access to address spaces of adapter
US8769180B2 (en) 2010-06-23 2014-07-01 International Business Machines Corporation Upbound input/output expansion request and response processing in a PCIe architecture
US8918573B2 (en) 2010-06-23 2014-12-23 International Business Machines Corporation Input/output (I/O) expansion response processing in a peripheral component interconnect express (PCIe) environment
US9298659B2 (en) 2010-06-23 2016-03-29 International Business Machines Corporation Input/output (I/O) expansion response processing in a peripheral component interconnect express (PCIE) environment
US9213661B2 (en) 2010-06-23 2015-12-15 International Business Machines Corporation Enable/disable adapters of a computing environment
US9134911B2 (en) 2010-06-23 2015-09-15 International Business Machines Corporation Store peripheral component interconnect (PCI) function controls instruction
US8645606B2 (en) 2010-06-23 2014-02-04 International Business Machines Corporation Upbound input/output expansion request and response processing in a PCIe architecture
US9201830B2 (en) 2010-06-23 2015-12-01 International Business Machines Corporation Input/output (I/O) expansion response processing in a peripheral component interconnect express (PCIe) environment
US9626298B2 (en) 2010-06-23 2017-04-18 International Business Machines Corporation Translation of input/output addresses to memory addresses
US9003084B2 (en) * 2011-02-18 2015-04-07 Ab Initio Technology Llc Sorting
US8762604B2 (en) 2011-02-18 2014-06-24 Ab Initio Technology Llc Managing buffer conditions through sorting
US8447901B2 (en) 2011-02-18 2013-05-21 Ab Initio Technology Llc Managing buffer conditions through sorting
US9128686B2 (en) 2011-02-18 2015-09-08 Ab Initio Technology Llc Sorting

Also Published As

Publication number Publication date Type
US7496707B2 (en) 2009-02-24 grant
CN101132361A (en) 2008-02-27 application
US20080052441A1 (en) 2008-02-28 application

Similar Documents

Publication Publication Date Title
US7398335B2 (en) Method and system for DMA optimization in host bus adapters
US7530071B2 (en) Facilitating access to input/output resources via an I/O partition shared by multiple consumer partitions
US7073022B2 (en) Serial interface for a data storage array
US8060775B1 (en) Method and apparatus for providing dynamic multi-pathing (DMP) for an asymmetric logical unit access (ALUA) based storage system
US20090100194A1 (en) System and method of emulating a network controller within an information handling system
US20030099254A1 (en) Systems and methods for interfacing asynchronous and non-asynchronous data media
US20040068673A1 (en) System and method for managing the operating frequency of processors or blades
US6760783B1 (en) Virtual interrupt mechanism
US20060047852A1 (en) Method and system for using boot servers in networks
US7197591B2 (en) Dynamic lane, voltage and frequency adjustment for serial interconnect
US20060259656A1 (en) Simulating multiple virtual channels in switched fabric networks
US20080104283A1 (en) Method and system for achieving fair command processing in storage systems that implement command-associated priority queuing
US20040179534A1 (en) Method and apparatus for shared I/O in a load/store fabric
US20050102437A1 (en) Switching apparatus and method for link initialization in a shared I/O environment
US7610431B1 (en) Configuration space compaction
US7602774B1 (en) Quality of service for server applications
US20060280195A1 (en) Systems and methods for providing dedicated or shared network interface functionality via a single MAC
US20110078679A1 (en) Provisioning virtual machine placement
US20060168377A1 (en) Reallocation of PCI express links using hot plug event
US20050027900A1 (en) Method and apparatus for a shared I/O serial ATA controller
US20110106981A1 (en) Configuration space compaction
US20120054264A1 (en) Techniques for Migrating Active I/O Connections with Migrating Servers and Clients
US20080005410A1 (en) Methodology for manipulation of SATA device access cycles
US20070050538A1 (en) Smart scalable storage switch architecture
US20080141264A1 (en) Methods and systems for load balancing of virtual machines in clustered processors using storage related load information