US20060048160A1 - Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor - Google Patents

Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor Download PDF

Info

Publication number
US20060048160A1
US20060048160A1 US10/932,702 US93270204A US2006048160A1 US 20060048160 A1 US20060048160 A1 US 20060048160A1 US 93270204 A US93270204 A US 93270204A US 2006048160 A1 US2006048160 A1 US 2006048160A1
Authority
US
United States
Prior art keywords
idle
processor
time
thread
length
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
US10/932,702
Inventor
Bret Olszewski
Randal Swanberg
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
Application filed by International Business Machines Corp filed Critical International Business Machines Corp
Priority to US10/932,702 priority Critical patent/US20060048160A1/en
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: SWANBERG, RANDAL CRAIG, OLSZEWSKI, BRET RONALD
Assigned to INTERNATIONAL BUSINESS MACHINES CORPORATION reassignment INTERNATIONAL BUSINESS MACHINES CORPORATION ASSIGNMENT OF ASSIGNORS INTEREST (SEE DOCUMENT FOR DETAILS). Assignors: SWANBERG, RANDAL CRAIG, OLSZEWSKI, BRET RONALD
Publication of US20060048160A1 publication Critical patent/US20060048160A1/en
Application status is Abandoned legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083Techniques for rebalancing the load in a distributed system
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F2209/00Indexing scheme relating to G06F9/00
    • G06F2209/50Indexing scheme relating to G06F9/50
    • G06F2209/508Monitor

Abstract

A method, apparatus, and computer program product are disclosed for dynamically determining when to yield a processor that is assigned to perform particular work but that is currently idle. A particular processor is assigned to perform work. A determination is made regarding whether the processor is currently idle. If the processor is currently idle, a determination is made of a length of time the processor has been idle. If this determined length of time exceeds a self-tunable threshold, the processor is yielded to make the processor available to perform other work. The threshold can be dynamically tuned during runtime.

Description

    BACKGROUND OF THE INVENTION
  • 1. Technical Field
  • The present invention is directed to data processing systems. More specifically, the present invention is directed to a method, apparatus, and computer program product for utilizing a self-tuning parameter to dynamically determine when to yield a processor that is assigned to perform particular work but that is currently idle.
  • 2. Description of Related Art
  • A symmetric multiprocessing (SMP) data processing system has multiple processors that are symmetric such that each processor has the same processing speed and latency. An SMP system has one operating system that divides the work into tasks that are distributed evenly among the various processors by dispatching one software thread of work to each processor at a time. Thus, a processor in an SMP system executes only one thread at a time.
  • A simultaneous multi-threading (SMT) data processing system includes multiple processors that can each concurrently execute more than one thread at a time per processor. An SMT system has the ability to favor one thread over another when both threads are running on the same processor. If one SMT thread is idle, the performance of other SMT threads that are executing on the same core can be improved by having the idle SMT thread yield its portion of its processor's resources.
  • Known systems can include a shared processor where the shared processor is shared among the various processes that are being executed by the system. A shared processor may be part of a logically partitioned system and shared among the various partitions in the system. These systems typically include firmware, also called a hypervisor, that manages and enforces the partitioning and/or sharing of the processor. For example, a hypervisor may receive a request from the system to dispatch a virtual processor to a physical processor. The virtual processor includes a definition of the work to be done by a physical processor as well as various settings and state information that are required to be set within the physical processor in order for the physical processor to execute the work.
  • In known systems there exists a pool of one or more available physical processors. When in the pool, these processors are available and currently unassigned. The pool is managed by a process or firmware such as a hypervisor. Any time work is needed to be performed, a request will be received by the hypervisor to dispatch a particular virtual processor that defines the work. The hypervisor then selects a physical processor from the pool to use to perform this work to which the virtual processor will be dispatched. When the physical processor is selected, it is logically removed from the pool and temporarily assigned to perform the work defined by the virtual processor. This temporary assignment is done when a particular virtual processor is dispatched by the hypervisor to a particular physical processor.
  • When a virtual processor is dispatched to a physical processor, the physical processor is assigned to that virtual processor for a preset period of time. The physical processor is assigned for that period of time regardless of the amount of time that may be needed to complete the work defined by the virtual processor. Thus, when the physical processor finishes the work defined by the virtual processor before the preset period of time expires, the physical processor will not have any useful work to do but will remain assigned to that virtual processor. In this case, the physical processor will be idle until the time period ends, i.e. it is not performing any useful work until the time period ends. When the time period ends, the physical processor is released from this temporary assignment and is again logically included in the pool of available processors.
  • Another example of a processor waiting for work is when a process being executed by the processor is spinning, waiting to acquire a software lock. Software locks are known in the art. Locks are typically used to serialize access to some shared resource. Thus, prior to a thread accessing a shared resource, it is required that the thread acquire a particular software lock. After a thread acquires the lock, that thread holds the lock as long as it requires access to the shared resource. Other threads desiring to access the same resource must wait until the thread finishes its work and releases the lock. When the lock is released, it is available. Waiting or spinning threads may again contend for the lock. Thus, when a process is spinning, the processor that is executing the process is not able to perform other useful work.
  • It is known in the art to be able to yield an idle processor. When a processor is yielded, it becomes available to be assigned to do other work. When it is yielded, it becomes part of the pool of available processors.
  • Therefore, a need exists for a method, apparatus, and computer program product for providing a self-tuning parameter that is used to dynamically determine when to yield a processor that is assigned to perform particular work but that is currently idle.
  • SUMMARY OF THE INVENTION
  • A method, apparatus, and computer program product are disclosed for dynamically determining when to yield a processor that is assigned to perform particular work but that is currently idle. A particular processor is assigned to perform work. A determination is made regarding whether the processor is currently idle. If the processor is currently idle, a length of time is determined that the processor has been idle. If this determined length of time exceeds a self-tunable threshold, the processor is yielded to make the processor available to perform other work.
  • The threshold can be dynamically tuned during runtime. Statistics are collected during runtime regarding the time spent by processors in the system idling. The threshold can be dynamically tuned during runtime using the statistics without user intervention. The threshold can be adjusted to a new value where the new value is determined using the statistics.
  • 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
  • The novel features believed characteristic of the invention are set forth in the appended claims. The invention itself, however, as well as a preferred mode of use, further objectives 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:
  • FIG. 1 is a block diagram of a data processing system, including a hypervisor, in which the present invention may be implemented;
  • FIG. 2 is a more detailed illustration of a computer system that may be used to implement any of the computer systems described herein in which the present invention may be implemented in accordance with the present invention;
  • FIG. 3 is a block diagram of a logically partitioned platform that includes the present invention;
  • FIGS. 4A and 4B together depict a high level flow chart that illustrates in more detail using a self-tunable parameter to determine when to yield a processor that is assigned to perform particular work but that is currently idle in accordance with the present invention; and
  • FIG. 5 illustrates a high level flow chart that depicts another example of utilizing a tunable parameter for determining when to yield a processor in accordance with the present invention.
  • DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT
  • A preferred embodiment of the present invention and its advantages are better understood by referring to the figures, like numerals-being used for like and corresponding parts of the accompanying figures.
  • The present invention is a method, apparatus, and computer program product for providing a self-tunable parameter that is used to dynamically determine when to yield a processor that is assigned to perform particular work but that is currently idle.
  • System throughput and performance can be increased by proactively yielding processors when the processors are idle. Each time a physical processor becomes available due to, for example, one virtual processor yielding its physical processor to be available to another virtual processor, a context switch must occur to switch the settings and state of the physical processor from the first virtual processor to the new virtual processor. While yielding an idle processor frees up the processor in a shared processor LPAR system, or boosts the performance of sibling threads in an SMT system, the net result is an increased latency for starting new work which can noticeably impact some workloads. By providing a tunable parameter that is used to control how quickly a shared processor or SMT thread will give up their resource for the benefit of another processor or SMT thread, the present invention will boost the performance of the system while reducing the latency impact for starting new work.
  • The present invention provides a tunable threshold parameter. This tunable parameter is used to determine how long an idle virtual processor will wait for work before yielding its physical processor to make the physical processor available to be claimed by another virtual processor. This parameter can be set to “0”, meaning that the physical processor will always be yielded when idle, be set to “infinity” meaning that the physical processor will never be yielded when idle, or be set to some value in between.
  • Typically when a virtual processor is dispatched to a physical processor, a particular physical processor is temporarily assigned and dedicated to that virtual processor for a particular period of time. This particular period of time is typically a set length of time that is defined within the system. When the period of time expires, regardless of whether the physical processor has finished executing the work defined by the virtual processor, the physical processor will rejoin the pool of available physical processors. In many cases, a physical processor will finish executing the work defined by a virtual processor before the period of time expires. In these cases, the physical processor is idle until the period of time expires.
  • The parameter can be modified as the system is processing work. This provides the ability to configure the system for maximum performance of the currently scheduled work by making all processing capacity available, for maximizing the performance of starting new work by avoiding latency, or for a reasonable balance to avoid latency for certain arrival rates of new work while donating processing power after some preset amount of delay.
  • FIG. 1 is a block diagram of a data processing system 100, including a hypervisor, in which the present invention may be implemented. System 100 is preferably a symmetric multiprocessing (SMP) server computer system. SMP server computer system 100 includes physical hardware devices that can be mapped to, i.e. temporarily owned by, a user application to execute that application.
  • SMP server computer system 100 includes a physical SMP server 102. Physical SMP server 102 includes physical hardware devices such as processors 104, memory 106, and I/O adapters 108. These physical devices are managed by hypervisor 110. Each one of the processors 104 is preferably a simultaneous multithreaded (SMT) processor that is capable of concurrently executing multiple different threads on the processor.
  • A virtual server is a proxy for a physical server that has the same capabilities, interfaces, and state. Virtual servers are created and managed by a hypervisor that resides on physical SMP server computer system 100. A virtual server appears to be a physical SMP server to its user: the operating system, middleware, and application software that run upon it. SMP server computer system 100 includes one or more virtual servers such as virtual server 112 and virtual server 112 a.
  • Each virtual server appears to its software to include its own processor(s), memory, and I/O adapter(s) that are available for the exclusive use of that virtual server. For example, virtual server 112 includes a virtual processor 120, virtual memory 122, and virtual I/O adapters 124. Virtual server 112 a includes virtual processors 120 a, virtual memory 122 a, and virtual I/O adapters 124 a.
  • Each virtual server supports its own software environment, including an operating system, middleware, and applications. The software environment of each virtual server can be different from the software environment of other virtual servers. For example, the operating systems executed by each virtual server may differ from one another.
  • For example, virtual server 112 supports operating system 114, middleware 116, and applications 118. Virtual server 112 a supports operating system 114 a, middleware 116 a, and applications 118 a. Operating systems 114 and 114 a may be the same or different operating systems.
  • A virtual server is a logical description of a server that defines a server environment that acts, to a user, as if it were a physical server, being accessed and providing information in the same way as a physical server. The virtual processors, virtual memory, and virtual I/O adapters that are defined for each virtual server are logical substitutes for physical processors, memory, and I/O adapters.
  • Hypervisor 110 manages the mapping between the virtual servers with their virtual processors, virtual memory, and virtual I/O adapters and the physical hardware devices that are selected to implement these virtual devices. For example, when a virtual processor is dispatched, a physical processor, such as one of physical processors 104, is selected by hypervisor 110 to be used to execute and implement that virtual processor. Hypervisor 110 manages the selections of physical devices and their temporary assignment to virtual devices.
  • Hypervisor 110 is responsible for dynamically creating, managing, and destroying virtual SMP servers. Whole virtual processors, virtual I/O adapters, and virtual memory blocks can be removed or added by hypervisor 110. Hypervisor 110 is also responsible for dynamic resource allocation, managing time-sharing of physical resources, and altering the physical resource mapped to a processor without involving the operating system. Hypervisor 110 is also able to dedicate physical resources to virtual resources for situations where sharing is not desired. Hypervisor 110 is responsible for managing the addition or removal of physical resources. Hypervisor 110 makes these additions and deletions transparent to the upper level applications.
  • FIG. 2 is a more detailed illustration of a computer system that may be used to implement any of the computer systems described herein in which the present invention may be implemented in accordance with the present invention. Data processing system 200 may be a symmetric multiprocessor (SMP) system including a plurality of SMT-capable processors 202 and 204 connected to system bus 206. Alternatively, a single processor system may be employed. In the depicted example, processor 204 is a service processor. All of the processors described herein with reference to all of the figures are SMT-capable processors. Each SMT-capable processor is capable of concurrently executing multiple threads on the one processor.
  • Also connected to system bus 206 is memory controller/cache 208, which provides an interface to local memory 209. I/O bus bridge 210 is connected to system bus 206 and provides an interface to I/O bus 212. Memory controller/cache 208 and I/O bus bridge 210 may be integrated as depicted.
  • Peripheral component interconnect (PCI) bus bridge 214 connected to I/O bus 212 provides an interface to PCI local bus 216. A number of modems may be connected to PCI bus 216. Typical PCI bus implementations will support four PCI expansion slots or add-in connectors. Communications links to network computers 108-112 in FIG. 1 may be provided through modem 218 and network adapter 220 connected to PCI local bus 216 through add-in boards.
  • Network adapter 220 includes a physical,layer 282 which conditions analog signals to go out to, the network, such as for example an Ethernet network over an R45 connector. A media access controller (MAC) 280 is included within network adapter 220. Media access controller (MAC) 280 is coupled to bus 216 and processes digital network signals. MAC 280 serves as an interface between bus 216 and physical layer 282. MAC 280 performs a number of functions involved in the transmission and reception of data packets. For example, during the transmission of data, MAC 280 assembles the data to be transmitted into a packet with address and error detection fields. Conversely, during the reception of a packet, MAC 280 disassembles the packet and performs address checking and error detection. In addition, MAC 280 typically performs encoding/decoding of digital signals transmitted and performs preamble generation/removal as well as bit transmission/reception.
  • Additional PCI bus bridges 222 and 224 provide interfaces for additional PCI buses 226 and 228, from which additional modems or network adapters may be supported. In this manner, data processing system 200 allows connections to multiple network computers. A memory-mapped graphics adapter 230 and hard disk 232 may also be connected to I/O bus 212 as depicted, either directly or indirectly.
  • Service processor 204 interrogates system processors, memory components, and I/O bridges to generate an inventory and topology understanding of data processing system 200. Service processor 204 also executes Built-In-Self-Tests (BISTs), Basic Assurance Tests (BATs), and memory tests on all elements found by interrogating a system processor, memory controller, and I/O bridge. Any error information for failures detected during the BISTs, BATs, and memory tests are gathered and reported by service processor 204.
  • Those of ordinary skill in the art will appreciate that the hardware depicted in FIG. 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. The depicted example is not meant to imply architectural limitations with respect to the present invention.
  • The present invention may be executed by one of the computers depicted by FIG. 1 or 2.
  • FIG. 3 is a block diagram of a logically partitioned platform that includes the present invention. Data processing system 320 includes logically partitioned platform 350. Platform 350 includes partitioned hardware 352, partition management firmware, also called a hypervisor 354, and partitions 356-359. Operating systems 361-364 exist within partitions 356-359. Operating systems 361-364 may be multiple copies of a single operating system or multiple heterogeneous operating systems simultaneously run on platform 350.
  • Partitioned hardware 352 includes a plurality of SMT-capable processors 365-368, a plurality of system memory units 370-373, a plurality of input/output (I/O) adapters 374-381, and a storage unit 382. Each of the processors 365-368, memory units 370-373, NVRAM storage 383, and I/O adapters 374-381.may be assigned to one of multiple partitions 356-359. Partitioned hardware 352 also includes service processor 390. A non-volatile memory device 391, such as an NVRAM device, is included within service processor 390.
  • Partition management firmware (hypervisor) 354 performs a number of functions and services for partitions 356-359 to create and enforce the partitioning of logically partitioned platform 350. Hypervisor 354 is a firmware implemented virtual machine identical to the underlying hardware. Firmware is “software” stored in a memory chip that holds its content without electrical power, such as, for example, read-only memory (ROM), programmable ROM (PROM), erasable programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), and non-volatile random access memory (non-volatile RAM). Thus, hypervisor 354 allows the simultaneous execution of independent OS images 361-364 by virtualizing all the hardware resources of logically partitioned platform 250. Hypervisor 354 may attach I/O devices through I/O adapters 374-381 to single virtual machines in an exclusive mode for use by one of OS images 361-364.
  • A hardware management console (HMC) 380 may be coupled to service processor 390 in data processing system 320. HMC 380 is a separate computer system that is coupled to service processor 390 and may be used by a user to control various functions of system 320 through service processor 390.
  • FIGS. 4A and 4B together depict a high level flow chart that illustrates in more detail using a tunable parameter to determine when to yield a processor that is assigned to perform particular work but that is currently idle and that illustrates modifying the tunable parameter in accordance with the present invention. Block 400 depicts the start of a particular Idle process such as an idle thread. This idle thread is executed by a physical processor that has been allocated to a particular process, such as by having a virtual processor dispatched to the physical processor or by an SMT physical processor executing a thread. It is known in the art for a processor to select and execute an Idle thread when that processor is not currently performing any work. Thus, when a processor is waiting for work to do, the processor will execute the Idle thread. The process depicted by FIGS. 4A and 4B are executed by the idle thread. Next, block 402 illustrates setting a Start_Idle_Timestamp equal to a Current_Timestamp. The Current_Timestamp is the current value of the system's clock. Thus, the Current_Timestamp represents the current real time. The Start_Idle_Timestamp is used to indicate when a processor first started idling. This indicates the start of an interval.
  • The process then passes to block 404 which depicts setting the Idle_Flag equal to “TRUE”. The Idle_Flag is used to indicate whether the processor is currently idling. If the processor is idling, the Idle_Flag will be set to “TRUE”. When the processor is not idling, the Idle_Flag will be set to “FALSE”.
  • Next, block 406 illustrates initializing the Idle_Limit parameter to be equal to a default value. The Idle_Limit parameter is a self-tunable threshold that is used to determine when to yield a processor that is idle. During runtime and without user intervention, the Idle_Limit parameter can be adjusted to a new value that is determined using the average time spent idling statistics.
  • Block 408, then, illustrates a determination of whether or not the Idle_Flag is equal to “TRUE”. If a determination is made that the Idle_Flag is not equal to “TRUE”, i.e., it is equal to “FALSE”, the process passes to block 410 which depicts setting the Idle_Interval equal to the End_Idle_Timestamp minus the Start_Idle_Timestamp. The End_Idle_Timestamp is set when the processor is no longer idle. The Idle_Interval value is the amount of time this processor was idle during this entire interval. When the process passes to block 410, an interval has occurred. This interval is an interval between a start and an end sample. The start of a sample is marked by the Start_Idle_Timestamp, and the end of that sample is marked by the End_Idle_Timestamp. Next, block 412 illustrates adding the Idle_Interval value to the Idle_Accumulator value and incrementing the current number of intervals. The Idle_Accumulator value is the total amount of time that all processors have been idle during all intervals. Block 414, then, depicts resetting the Start_Idle_Timestamp to be equal to the Current_Timestamp in order to mark the beginning of the next sample. The Idle_Flag is set equal to “TRUE”.
  • The process then passes to block 416 which illustrates computing the average time spent idling by all processors in the system by dividing the Idle_Accumulator value by the number of intervals. The number of intervals is another variable that is also maintained by the present invention. The number of intervals is incremented each time the End_Idle_Timestamp is set equal to the Current_Timestamp. Thereafter, block 418 depicts a determination of whether or not the system is enabled for autonomic yield tuning. This is a determination of whether or not the system is enabled to modify the Idle_Limit parameter itself during operation without any user input. If a determination is made that the system is not enabled for autonomic yield tuning, the process passes to block 408. If a determination is made that the system is enabled for autonomic yield tuning, the Idle threshold, represented by the Idle_Limit value, can dynamically adjust itself during runtime. If the system is enabled for autonomic yielding, the process passes to block 420 which depicts setting the Idle_Limit value equal to the average time spent idling determined as depicted by block 416. Therefore, the Idle_Limit threshold is dynamically adjusted using the average time spent idling statistics. The process then passes back to block 408.
  • There are several methods for reentering or resuming the same particular idle process in a particular processor that had been suspended. The same idle process is defined as being the idle process that is executed by a particular processor when that physical processor has no more work to do. This idle process may be interrupted and then later resumed. For example, an I/O interrupt to the processor currently running the idle process can cause the idle process to be interrupted while useful work occurs in the interrupt handler. Another example is when the idle process yields its processor, the idle process is interrupted. When new work becomes available for the idle virtual processor, the idle virtual processor can be “prodded” such that its idle process is resumed and will then see the new work available. The present invention tracks the amount of time the processor spent idling before new work became available for it. That new work could be defined as interrupt processing work, an existing sleeping job becoming runnable again, or a new job ready to run for the first time.
  • One method for resuming or reentering the same idle process that had been suspended is depicted by block 400 a. Block 400 a depicts this same idle process being redispatched by an operating system when there is no more work for this processor to perform. When the same idle process is redispatched by an operating system when there is no more work for this processor to perform, the process passes back to block 408.
  • Another method for resuming or reentering the same idle process that had been suspended in a particular processor is illustrated by block 400 b. Block 400 b depicts either a virtual processor being redispatched by the hypervisor to this processor, or an SMT thread being redispatched to this processor by the hypervisor. One of these events will occur when another processor revived this processor. When another processor revives this processor, it is because there is now useful work for this processor to perform. In either case, if the hypervisor redispatched a virtual processor or an SMT thread to this processor, the process passes to block 424 which illustrates setting the End_Idle_Timestamp equal to the Current_Timestamp. The Idle_Flag is then set to “FALSE”. The process then passes to block 408.
  • Another method for resuming or reentering the same idle process that had been suspended in a particular processor is illustrated by block 400 c. Block 400 c depicts either a virtual processor being redispatched by the hypervisor to this processor, or an SMT thread being redispatched to this processor by the hypervisor due to a clock or I/O interrupt. The hypervisor will redispatch the virtual processor or SMT thread when there is work for the processor to do. Thus, by definition, when the hypervisor redispatches the virtual processor or SMT thread, the processor has been assigned work to do and is therefore no longer idle. If the hypervisor redispatched a virtual processor or an SMT thread to this processor due to a clock or I/O interrupt, the process passes to block 424 and thereafter to block 408.
  • Referring again to block 408, if a determination is made that the Idle_Flag is equal to “TRUE”, the process passes to block 430 (see FIG. 4B). Block 430 depicts setting the Current_Idle value equal to the Current_Timestamp minus the Start_Idle_Timestamp. Next, block 432 illustrates a determination of whether or not the Current_Idle value is greater than the Idle_Limit. If a determination is made that the Current_Idle value is greater than the Idle_Limit, the process passes to block 434 which depicts yielding this processor. An SMT processor can be yielded, for example, by transitioning the processor into single threaded mode. A shared processor can be yielded, for example, by ceding the virtual processor to the hypervisor for redispatch.
  • The idle process is then exited as illustrated by block 436. Referring again to block 432, if a determination is made that the Current_Idle value is not greater than the Idle_Limit, the process passes to block 438 which depicts a determination of whether or not there is any available work for this processor to claim. If a determination is made that there is no available work for this processor to claim, the process passes back to block 408 (see FIG. 4A).
  • Referring again to block 438, if there is available work for this processor to claim, the process passes to block 440 which illustrates setting the End_Idle_Timestamp equal to the Current_Timestamp. The Idle_Flag is set equal to “FALSE”. Next, block 442 depicts this processor claiming the available work by calling the operating system dispatcher to dispatch the newly claimed work. This idle process is then either exited or suspended as appropriate as illustrated by block 436.
  • FIG. 5 illustrates a high level flow chart that depicts another example of utilizing a tunable parameter for determining when to yield a processor in accordance with the present invention. The process starts as depicted by block 500 and thereafter passes to block 502 which illustrates clearing an accumulator of elapsed time spent spinning waiting to acquire a software lock. Next, block 504 depicts a process that is being executed by a processor trying to acquire the lock. Block 506, then, illustrates a determination of whether or not the process was able to acquire the software lock. If a determination is made that the process was able to acquire the software lock, the process passes to block 508 which depicts exiting this “spin lock” process.
  • Referring again to block 506, if a determination is made that the process was not able to acquire the software lock, the process passes to block 510 which illustrates accumulating in the accumulator the time spent spinning by this process waiting to acquire the lock. Next, block 512 depicts determining the total time spent so far by this process spinning waiting to acquire the lock. Block 514, then, illustrates a determination of whether or not the total time spent by the process spinning is greater than the Idle threshold. If a determination is made that the total time spent spinning is not greater than the Idle threshold, the process passes back to block 504.
  • Referring again to block 514, if a determination is made that the total time spent spinning is greater than the Idle threshold, the process passes to block 516 which illustrates yielding this processor. The process then passes to block 508 which depicts exiting this “spin lock” process.
  • As described above with reference to the Idle threshold that was utilized for yielding a processor, the Idle threshold used for determining if a spin lock process has spent too long spinning is also autonomic, or self-tunable. The spin lock process Idle threshold could tune itself, similarly to the yielding process Idle threshold, by setting a new spin lock process Idle threshold to a time spent spinning waiting for a lock.
  • It is important to note that while the present invention has been described in the context of a fully functioning data processing system. Those of ordinary skill in the art will appreciate that the processes of the present invention are capable of being distributed in the form of a computer readable medium of instructions and a variety of forms and that the present invention applies equally regardless of the particular type of signal bearing media actually used to carry out the distribution. Examples of computer readable media include recordable-type media, such as a floppy disk, a hard disk drive, a RAM, CD-ROMs, DVD-ROMs, and transmission-type media, such as digital and analog communications links, wired or wireless communications links using transmission forms, such as, for example, radio frequency and light wave transmissions. The computer readable media may take the form of coded formats that are decoded for actual use in a particular data processing system.
  • The description of the present invention has been presented for purposes of illustration and description, and is not intended to be exhaustive or limited to the invention in the form disclosed. Many modifications and variations will be apparent to those of ordinary skill in the art. The embodiment was chosen and described in order to best explain the principles of the invention, the practical application, and to enable others of ordinary skill in the art to understand the invention for various embodiments with various modifications as are suited to the particular use contemplated.

Claims (20)

1. A method in a shared processor data processing system for dynamically determining when to yield a processor that is assigned to perform particular work but that is currently idle, said method comprising:
assigning a particular processor to perform work;
determining whether said processor is currently idle;
in response to determining whether said processor is currently idle, determining a length of time said processor has been idle; and
in response to said length of time exceeding a threshold, yielding said processor to make said processor available to perform other work.
2. The method according to claim 1, further comprising:
self-tuning said threshold during runtime without user intervention by autonomically adjusting said threshold to a new value.
3. The method according to claim 2, further comprising:
said system including a plurality of processors;
determining an average length of time all of said plurality of processors are idle; and
self-tuning said threshold by setting said threshold equal to said average length of time.
4. The method according to claim 1, further comprising:
executing an idle thread by said processor when said processor is idle;
accumulating, by said idle thread, said length of time said processor has been idle;
determining, by said idle thread, said length of time said processor has been idle; and
determining, by said idle thread, whether said length of time exceeds said threshold.
5. The method according to claim 4, further comprising:
determining, by said idle thread, an average length of time all of said plurality of processors are idle; and
self-tuning, by said idle thread, said threshold by setting said threshold equal to said average length of time.
6. The method according to claim 4, further comprising: temporarily suspending said execution of said idle thread while said processor executes work;
pausing said accumulation, by said idle thread, of said length of time said processor has been idle when said idle thread execution is suspended;
resuming said execution of said idle thread;
resuming said accumulation, by said idle thread, of said length of time said processor has been idle when said idle thread execution is resumed.
7. The method according to claim 6, further comprising:
exiting said idle thread;
determining a start of an interval when said processor first becomes idle; and
determining an end of said interval when said idle thread is exited.
8. An apparatus in a shared processor data processing system for dynamically determining when to yield a processor that is assigned to perform particular work but that is currently idle, said apparatus comprising:
a particular processor assigned to perform work;
said system including firmware for determining whether said processor is currently idle;
in response to determining whether said processor is currently idle, an idle thread for determining a length of time said processor has been idle; and
in response to said length of time exceeding a threshold, said firmware for yielding said processor to make said processor available to perform other work.
9. The apparatus according to claim 8, further comprising:
said threshold being self-tuned during runtime without user intervention by autonomically adjusting said threshold to a new value.
10. The apparatus according to claim 9, further comprising:
said system including a plurality of processors;
said idle thread for determining an average length of time all of said plurality of processors are idle; and
said idle thread for self-tuning said threshold by setting said threshold equal to said average length of time.
11. The apparatus according to claim 8, further comprising:
said processor executing an idle thread when said processor is idle;
said idle thread accumulating said length of time said processor has been idle;
said idle thread determining said length of time said processor has been idle; and
said idle thread determining whether said length of time exceeds said threshold.
12. The apparatus according to claim 11, further comprising:
said idle thread determining an average length of time all of said plurality of processors are idle; and
said idle thread self-tuning said threshold by setting said threshold equal to said average length of time.
13. The apparatus according to claim 11, further comprising:
said firmware temporarily suspending said execution of said idle thread while said processor executes work;
said idle thread pausing said accumulation of said length of time said processor has been idle when said idle thread execution is suspended;
said firmware resuming said execution of said idle thread;
said idle thread resuming said accumulation of said length of time said processor has been idle when said idle thread execution is resumed.
14. The apparatus according to claim 13, further comprising:
said processor exiting said idle thread;
said idle thread determining a start of an interval when said processor first becomes idle; and
said idle thread determining an end of said interval when said idle thread is exited.
15. A computer program product in a shared processor data processing system for dynamically determining when to yield a processor that is assigned to perform particular work but that is currently idle, said product comprising:
instructions for assigning a particular processor to perform work;
instructions for determining whether said processor is currently idle;
in response to determining whether said processor is currently idle, instructions for determining a length of time said processor has been idle; and
in response to said length of time exceeding a threshold, instructions for yielding said processor to make said processor available to perform other work.
16. The product according to claim 15, further comprising:
instructions for self-tuning said threshold during runtime without user intervention by autonomically adjusting said threshold to a new value.
17. The product according to claim 16, further comprising:
said system including a plurality of processors;
instructions for determining an average length of time all of said plurality of processors are idle; and
instructions for self-tuning said threshold by setting said threshold equal to said average length of time.
18. The product according to claim 15, further comprising:
instructions for executing an idle thread by said processor when said processor is idle;
instructions for accumulating, by said idle thread, said length of time said processor has been idle;
instructions for determining, by said idle thread, said length of time said processor has been idle; and
instructions for determining, by said idle thread, whether said length of time exceeds said threshold.
19. The product according to claim 18, further comprising:
instructions for temporarily suspending said execution of said idle thread while said processor executes work;
instructions for pausing said accumulation, by said idle thread, of said length of time said processor has been idle when said idle thread execution is suspended;
instructions for resuming said execution of said idle thread;
instructions for resuming said accumulation, by said idle thread, of said length of time said processor has been idle when said idle thread execution is resumed.
20. The product according to claim 19, further comprising:
instructions for exiting said idle thread;
instructions for determining a start of an interval when said processor first becomes idle; and
instructions for determining an end of said interval when said idle thread is exited.
US10/932,702 2004-09-02 2004-09-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor Abandoned US20060048160A1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US10/932,702 US20060048160A1 (en) 2004-09-02 2004-09-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
US10/932,702 US20060048160A1 (en) 2004-09-02 2004-09-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor
US12/061,353 US8141083B2 (en) 2004-09-02 2008-04-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US12/061,353 Continuation US8141083B2 (en) 2004-09-02 2008-04-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor

Publications (1)

Publication Number Publication Date
US20060048160A1 true US20060048160A1 (en) 2006-03-02

Family

ID=35945013

Family Applications (2)

Application Number Title Priority Date Filing Date
US10/932,702 Abandoned US20060048160A1 (en) 2004-09-02 2004-09-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor
US12/061,353 Expired - Fee Related US8141083B2 (en) 2004-09-02 2008-04-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor

Family Applications After (1)

Application Number Title Priority Date Filing Date
US12/061,353 Expired - Fee Related US8141083B2 (en) 2004-09-02 2008-04-02 Method, apparatus, and computer program product for providing a self-tunable parameter used for dynamically yielding an idle processor

Country Status (1)

Country Link
US (2) US20060048160A1 (en)

Cited By (34)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20060075207A1 (en) * 2004-09-29 2006-04-06 Sony Corporation Information processing apparatus, process control method, and computer program
US20060143328A1 (en) * 2004-12-28 2006-06-29 Christian Fleischer Failover protection from a failed worker node in a shared memory system
US20060156306A1 (en) * 2004-12-13 2006-07-13 Infineon Technologies Ag Thread scheduling method, and device to be used with a thread scheduling method
US20070156869A1 (en) * 2005-12-30 2007-07-05 Galin Galchev Load balancing algorithm for servicing client requests
US20080163203A1 (en) * 2006-12-28 2008-07-03 Anand Vaijayanthimala K Virtual machine dispatching to maintain memory affinity
US20080271043A1 (en) * 2007-04-27 2008-10-30 Hyun Kim Accurate measurement of multithreaded processor core utilization and logical processor utilization
US20090199030A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Hardware Wake-and-Go Mechanism for a Data Processing System
US20090199184A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Wake-and-Go Mechanism With Software Save of Thread State
US20090199197A1 (en) * 2008-02-01 2009-08-06 International Business Machines Corporation Wake-and-Go Mechanism with Dynamic Allocation in Hardware Private Array
US20090199029A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Wake-and-Go Mechanism with Data Monitoring
US20090300317A1 (en) * 2008-05-30 2009-12-03 International Business Machines Corporation System and method for optimizing interrupt processing in virtualized environments
US20100185847A1 (en) * 2009-01-20 2010-07-22 New York University Database outsourcing with access privacy
US20100268790A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Complex Remote Update Programming Idiom Accelerator
US20100269115A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Managing Threads in a Wake-and-Go Engine
US20100268791A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Programming Idiom Accelerator for Remote Update
US20100293340A1 (en) * 2008-02-01 2010-11-18 Arimilli Ravi K Wake-and-Go Mechanism with System Bus Response
US20100293341A1 (en) * 2008-02-01 2010-11-18 Arimilli Ravi K Wake-and-Go Mechanism with Exclusive System Bus Response
US20110173423A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Look-Ahead Hardware Wake-and-Go Mechanism
US20110173417A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Programming Idiom Accelerators
US20110173419A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Look-Ahead Wake-and-Go Engine With Speculative Execution
US8127080B2 (en) 2008-02-01 2012-02-28 International Business Machines Corporation Wake-and-go mechanism with system address bus transaction master
US8171476B2 (en) 2008-02-01 2012-05-01 International Business Machines Corporation Wake-and-go mechanism with prioritization of threads
US8225120B2 (en) 2008-02-01 2012-07-17 International Business Machines Corporation Wake-and-go mechanism with data exclusivity
US20120227042A1 (en) * 2004-12-16 2012-09-06 Vmware, Inc. Mechanism for scheduling execution of threads for fair resource allocation in a multi-threaded and/or multi-core processing system
US8312458B2 (en) 2008-02-01 2012-11-13 International Business Machines Corporation Central repository for wake-and-go mechanism
US8341635B2 (en) 2008-02-01 2012-12-25 International Business Machines Corporation Hardware wake-and-go mechanism with look-ahead polling
US8516484B2 (en) 2008-02-01 2013-08-20 International Business Machines Corporation Wake-and-go mechanism for a data processing system
US20130232254A1 (en) * 2012-03-02 2013-09-05 Computenext Inc. Cloud resource utilization management
US8725992B2 (en) 2008-02-01 2014-05-13 International Business Machines Corporation Programming language exposing idiom calls to a programming idiom accelerator
US8732683B2 (en) 2008-02-01 2014-05-20 International Business Machines Corporation Compiler providing idiom to idiom accelerator
US20140259014A1 (en) * 2011-10-06 2014-09-11 Hitachi, Ltd. Virtual server processing control method, system, and virtual server processing control management server
US8880853B2 (en) 2008-02-01 2014-11-04 International Business Machines Corporation CAM-based wake-and-go snooping engine for waking a thread put to sleep for spinning on a target address lock
US8886919B2 (en) 2009-04-16 2014-11-11 International Business Machines Corporation Remote update programming idiom accelerator with allocated processor resources
US9513961B1 (en) * 2014-04-02 2016-12-06 Google Inc. Monitoring application loading

Families Citing this family (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8108879B1 (en) * 2006-10-27 2012-01-31 Nvidia Corporation Method and apparatus for context switching of multiple engines
JP5349072B2 (en) 2009-02-17 2013-11-20 パナソニック株式会社 Resource exclusion control method and resource exclusion control device
US8453122B2 (en) * 2009-11-10 2013-05-28 International Business Machines Corporation Symmetric multi-processor lock tracing
US8122167B1 (en) 2010-08-06 2012-02-21 International Business Machines Corporation Polling in a virtualized information handling system
US9158668B2 (en) 2012-06-27 2015-10-13 International Business Machines Corporation System and program product to allocate and reallocate CPU resources needed to utilize full available network adapter bandwidth capacity for logical partition migration

Citations (40)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5463775A (en) * 1991-06-10 1995-10-31 International Business Machines Corporation System and method for performing monitoring of resources in a data processing system in real time
US5691920A (en) * 1995-10-02 1997-11-25 International Business Machines Corporation Method and system for performance monitoring of dispatch unit efficiency in a processing system
US5696701A (en) * 1996-07-12 1997-12-09 Electronic Data Systems Corporation Method and system for monitoring the performance of computers in computer networks using modular extensions
US5761274A (en) * 1994-09-08 1998-06-02 Fujitsu Limited Method for regurating call congestion and ISDN exchanger for performing the same
US5774718A (en) * 1994-08-26 1998-06-30 Fujitsu Limited Process scheduling system that allocates CPU time with priority to interactive type processes
US5826079A (en) * 1996-07-05 1998-10-20 Ncr Corporation Method for improving the execution efficiency of frequently communicating processes utilizing affinity process scheduling by identifying and assigning the frequently communicating processes to the same processor
US5892959A (en) * 1990-06-01 1999-04-06 Vadem Computer activity monitor providing idle thread and other event sensitive clock and power control
US6049798A (en) * 1991-06-10 2000-04-11 International Business Machines Corporation Real time internal resource monitor for data processing system
US20020007387A1 (en) * 2000-06-02 2002-01-17 Michael Ginsberg Dynamically variable idle time thread scheduling
US6385638B1 (en) * 1997-09-04 2002-05-07 Equator Technologies, Inc. Processor resource distributor and method
US20020194250A1 (en) * 2001-06-18 2002-12-19 Bor-Ming Hsieh Sleep queue management
US20020194509A1 (en) * 2001-06-15 2002-12-19 Microsoft Corporation Method and system for using idle threads to adaptively throttle a computer
US6513057B1 (en) * 1996-10-28 2003-01-28 Unisys Corporation Heterogeneous symmetric multi-processing system
US20030033345A1 (en) * 2002-06-27 2003-02-13 Keefer Christopher E. Thread-based methods and systems for using the idle processing power of one or more networked computers to solve complex scientific problems
US20030041090A1 (en) * 2001-08-24 2003-02-27 Armstrong William Joseph Yield on multithreaded processors
US20030055864A1 (en) * 2001-08-24 2003-03-20 International Business Machines Corporation System for yielding to a processor
US20030191795A1 (en) * 2002-02-04 2003-10-09 James Bernardin Adaptive scheduling
US20030236816A1 (en) * 2002-06-20 2003-12-25 Lakshminarayanan Venkatasubramanian Spin-yielding in multi-threaded systems
US20040064817A1 (en) * 2001-02-28 2004-04-01 Fujitsu Limited Parallel process execution method and multiprocessor computer
US6728959B1 (en) * 1995-08-08 2004-04-27 Novell, Inc. Method and apparatus for strong affinity multiprocessor scheduling
US20040123297A1 (en) * 2002-11-12 2004-06-24 Arm Litmited Performance level setting of a data processing system
US6792601B1 (en) * 2000-05-18 2004-09-14 International Business Machines Corporation Multiple mode object locking method and system
US20040249947A1 (en) * 2003-05-22 2004-12-09 Hewlett-Packard Development Company, L.P. Concurrent cluster environment
US20050108711A1 (en) * 2003-11-13 2005-05-19 Infineon Technologies North America Corporation Machine instruction for enhanced control of multiple virtual processor systems
US20050108713A1 (en) * 2003-11-18 2005-05-19 Geye Scott A. Affinity mask assignment system and method for multiprocessor systems
US20050104799A1 (en) * 2003-11-14 2005-05-19 Shimizu Clifford S. Systems and methods for displaying individual processor usage in a multiprocessor system
US20050108714A1 (en) * 2003-11-18 2005-05-19 Geye Scott A. Dynamic resource management system and method for multiprocessor systems
US20050108398A1 (en) * 2003-11-14 2005-05-19 Microsoft Corporation Systems and methods for using metrics to control throttling and swapping in a message processing system
US20050108717A1 (en) * 2003-11-18 2005-05-19 Hong Steve J. Systems and methods for creating an application group in a multiprocessor system
US20050138443A1 (en) * 2003-12-23 2005-06-23 Barnes Cooper Method for C state management
US20050149929A1 (en) * 2003-12-30 2005-07-07 Vasudevan Srinivasan Method and apparatus and determining processor utilization
US20050188373A1 (en) * 2004-02-20 2005-08-25 Sony Computer Entertainment Inc. Methods and apparatus for task management in a multi-processor system
US20050229176A1 (en) * 2004-03-22 2005-10-13 Piotr Findeisen Determining processor usage by a thread
US6996822B1 (en) * 2001-08-01 2006-02-07 Unisys Corporation Hierarchical affinity dispatcher for task management in a multiprocessor computer system
US7043725B1 (en) * 1999-07-09 2006-05-09 Hewlett-Packard Development Company, L.P. Two tier arrangement for threads support in a virtual machine
US7043649B2 (en) * 2002-11-20 2006-05-09 Portalplayer, Inc. System clock power management for chips with multiple processing modules
US7131016B2 (en) * 2000-05-15 2006-10-31 Microconnect Llc Method and apparatus for adjusting clock throttle rate based on usage of CPU
US7162666B2 (en) * 2004-03-26 2007-01-09 Emc Corporation Multi-processor system having a watchdog for interrupting the multiple processors and deferring preemption until release of spinlocks
US7302684B2 (en) * 2001-06-18 2007-11-27 Microsoft Corporation Systems and methods for managing a run queue
US7343590B2 (en) * 2002-06-25 2008-03-11 Hewlett-Packard Development Company, L.P. System and method for increasing OS idle loop performance in a simulator

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7765543B1 (en) * 2003-12-17 2010-07-27 Vmware, Inc. Selective descheduling of idling guests running on a host computer system

Patent Citations (42)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5892959A (en) * 1990-06-01 1999-04-06 Vadem Computer activity monitor providing idle thread and other event sensitive clock and power control
US20020004912A1 (en) * 1990-06-01 2002-01-10 Amphus, Inc. System, architecture, and method for logical server and other network devices in a dynamically configurable multi-server network environment
US5572672A (en) * 1991-06-10 1996-11-05 International Business Machines Corporation Method and apparatus for monitoring data processing system resources in real-time
US5463775A (en) * 1991-06-10 1995-10-31 International Business Machines Corporation System and method for performing monitoring of resources in a data processing system in real time
US6049798A (en) * 1991-06-10 2000-04-11 International Business Machines Corporation Real time internal resource monitor for data processing system
US5774718A (en) * 1994-08-26 1998-06-30 Fujitsu Limited Process scheduling system that allocates CPU time with priority to interactive type processes
US5761274A (en) * 1994-09-08 1998-06-02 Fujitsu Limited Method for regurating call congestion and ISDN exchanger for performing the same
US6728959B1 (en) * 1995-08-08 2004-04-27 Novell, Inc. Method and apparatus for strong affinity multiprocessor scheduling
US5691920A (en) * 1995-10-02 1997-11-25 International Business Machines Corporation Method and system for performance monitoring of dispatch unit efficiency in a processing system
US5826079A (en) * 1996-07-05 1998-10-20 Ncr Corporation Method for improving the execution efficiency of frequently communicating processes utilizing affinity process scheduling by identifying and assigning the frequently communicating processes to the same processor
US5696701A (en) * 1996-07-12 1997-12-09 Electronic Data Systems Corporation Method and system for monitoring the performance of computers in computer networks using modular extensions
US6513057B1 (en) * 1996-10-28 2003-01-28 Unisys Corporation Heterogeneous symmetric multi-processing system
US6385638B1 (en) * 1997-09-04 2002-05-07 Equator Technologies, Inc. Processor resource distributor and method
US7043725B1 (en) * 1999-07-09 2006-05-09 Hewlett-Packard Development Company, L.P. Two tier arrangement for threads support in a virtual machine
US7131016B2 (en) * 2000-05-15 2006-10-31 Microconnect Llc Method and apparatus for adjusting clock throttle rate based on usage of CPU
US6792601B1 (en) * 2000-05-18 2004-09-14 International Business Machines Corporation Multiple mode object locking method and system
US20020007387A1 (en) * 2000-06-02 2002-01-17 Michael Ginsberg Dynamically variable idle time thread scheduling
US20040064817A1 (en) * 2001-02-28 2004-04-01 Fujitsu Limited Parallel process execution method and multiprocessor computer
US20020194509A1 (en) * 2001-06-15 2002-12-19 Microsoft Corporation Method and system for using idle threads to adaptively throttle a computer
US7302684B2 (en) * 2001-06-18 2007-11-27 Microsoft Corporation Systems and methods for managing a run queue
US20020194250A1 (en) * 2001-06-18 2002-12-19 Bor-Ming Hsieh Sleep queue management
US6996822B1 (en) * 2001-08-01 2006-02-07 Unisys Corporation Hierarchical affinity dispatcher for task management in a multiprocessor computer system
US20030055864A1 (en) * 2001-08-24 2003-03-20 International Business Machines Corporation System for yielding to a processor
US20030041090A1 (en) * 2001-08-24 2003-02-27 Armstrong William Joseph Yield on multithreaded processors
US20030191795A1 (en) * 2002-02-04 2003-10-09 James Bernardin Adaptive scheduling
US20030236816A1 (en) * 2002-06-20 2003-12-25 Lakshminarayanan Venkatasubramanian Spin-yielding in multi-threaded systems
US7343590B2 (en) * 2002-06-25 2008-03-11 Hewlett-Packard Development Company, L.P. System and method for increasing OS idle loop performance in a simulator
US20030033345A1 (en) * 2002-06-27 2003-02-13 Keefer Christopher E. Thread-based methods and systems for using the idle processing power of one or more networked computers to solve complex scientific problems
US20040123297A1 (en) * 2002-11-12 2004-06-24 Arm Litmited Performance level setting of a data processing system
US7043649B2 (en) * 2002-11-20 2006-05-09 Portalplayer, Inc. System clock power management for chips with multiple processing modules
US20040249947A1 (en) * 2003-05-22 2004-12-09 Hewlett-Packard Development Company, L.P. Concurrent cluster environment
US20050108711A1 (en) * 2003-11-13 2005-05-19 Infineon Technologies North America Corporation Machine instruction for enhanced control of multiple virtual processor systems
US20050104799A1 (en) * 2003-11-14 2005-05-19 Shimizu Clifford S. Systems and methods for displaying individual processor usage in a multiprocessor system
US20050108398A1 (en) * 2003-11-14 2005-05-19 Microsoft Corporation Systems and methods for using metrics to control throttling and swapping in a message processing system
US20050108713A1 (en) * 2003-11-18 2005-05-19 Geye Scott A. Affinity mask assignment system and method for multiprocessor systems
US20050108717A1 (en) * 2003-11-18 2005-05-19 Hong Steve J. Systems and methods for creating an application group in a multiprocessor system
US20050108714A1 (en) * 2003-11-18 2005-05-19 Geye Scott A. Dynamic resource management system and method for multiprocessor systems
US20050138443A1 (en) * 2003-12-23 2005-06-23 Barnes Cooper Method for C state management
US20050149929A1 (en) * 2003-12-30 2005-07-07 Vasudevan Srinivasan Method and apparatus and determining processor utilization
US20050188373A1 (en) * 2004-02-20 2005-08-25 Sony Computer Entertainment Inc. Methods and apparatus for task management in a multi-processor system
US20050229176A1 (en) * 2004-03-22 2005-10-13 Piotr Findeisen Determining processor usage by a thread
US7162666B2 (en) * 2004-03-26 2007-01-09 Emc Corporation Multi-processor system having a watchdog for interrupting the multiple processors and deferring preemption until release of spinlocks

Cited By (56)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US8286161B2 (en) * 2004-09-29 2012-10-09 Sony Corporation Apparatus, method, and computer program for setting and updating a relationship between a logical processor and a physical processor
US20060075207A1 (en) * 2004-09-29 2006-04-06 Sony Corporation Information processing apparatus, process control method, and computer program
US20060156306A1 (en) * 2004-12-13 2006-07-13 Infineon Technologies Ag Thread scheduling method, and device to be used with a thread scheduling method
US8108862B2 (en) * 2004-12-13 2012-01-31 Infineon Technologies Ag Out-of-order thread scheduling based on processor idle time thresholds
US10417048B2 (en) * 2004-12-16 2019-09-17 Vmware, Inc. Mechanism for scheduling execution of threads for fair resource allocation in a multi-threaded and/or multi-core processing system
US20120227042A1 (en) * 2004-12-16 2012-09-06 Vmware, Inc. Mechanism for scheduling execution of threads for fair resource allocation in a multi-threaded and/or multi-core processing system
US8140678B2 (en) 2004-12-28 2012-03-20 Sap Ag Failover protection from a failed worker node in a shared memory system
US20060143328A1 (en) * 2004-12-28 2006-06-29 Christian Fleischer Failover protection from a failed worker node in a shared memory system
US8707323B2 (en) * 2005-12-30 2014-04-22 Sap Ag Load balancing algorithm for servicing client requests
US20070156869A1 (en) * 2005-12-30 2007-07-05 Galin Galchev Load balancing algorithm for servicing client requests
US20080163203A1 (en) * 2006-12-28 2008-07-03 Anand Vaijayanthimala K Virtual machine dispatching to maintain memory affinity
US8024728B2 (en) 2006-12-28 2011-09-20 International Business Machines Corporation Virtual machine dispatching to maintain memory affinity
US8739162B2 (en) * 2007-04-27 2014-05-27 Hewlett-Packard Development Company, L.P. Accurate measurement of multithreaded processor core utilization and logical processor utilization
US20080271043A1 (en) * 2007-04-27 2008-10-30 Hyun Kim Accurate measurement of multithreaded processor core utilization and logical processor utilization
US8516484B2 (en) 2008-02-01 2013-08-20 International Business Machines Corporation Wake-and-go mechanism for a data processing system
US20100293340A1 (en) * 2008-02-01 2010-11-18 Arimilli Ravi K Wake-and-Go Mechanism with System Bus Response
US20100293341A1 (en) * 2008-02-01 2010-11-18 Arimilli Ravi K Wake-and-Go Mechanism with Exclusive System Bus Response
US20110173423A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Look-Ahead Hardware Wake-and-Go Mechanism
US20110173417A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Programming Idiom Accelerators
US20110173419A1 (en) * 2008-02-01 2011-07-14 Arimilli Ravi K Look-Ahead Wake-and-Go Engine With Speculative Execution
US8015379B2 (en) 2008-02-01 2011-09-06 International Business Machines Corporation Wake-and-go mechanism with exclusive system bus response
US8725992B2 (en) 2008-02-01 2014-05-13 International Business Machines Corporation Programming language exposing idiom calls to a programming idiom accelerator
US8732683B2 (en) 2008-02-01 2014-05-20 International Business Machines Corporation Compiler providing idiom to idiom accelerator
US8788795B2 (en) 2008-02-01 2014-07-22 International Business Machines Corporation Programming idiom accelerator to examine pre-fetched instruction streams for multiple processors
US8127080B2 (en) 2008-02-01 2012-02-28 International Business Machines Corporation Wake-and-go mechanism with system address bus transaction master
US8880853B2 (en) 2008-02-01 2014-11-04 International Business Machines Corporation CAM-based wake-and-go snooping engine for waking a thread put to sleep for spinning on a target address lock
US8145849B2 (en) 2008-02-01 2012-03-27 International Business Machines Corporation Wake-and-go mechanism with system bus response
US8640141B2 (en) 2008-02-01 2014-01-28 International Business Machines Corporation Wake-and-go mechanism with hardware private array
US20090199029A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Wake-and-Go Mechanism with Data Monitoring
US8225120B2 (en) 2008-02-01 2012-07-17 International Business Machines Corporation Wake-and-go mechanism with data exclusivity
US8640142B2 (en) 2008-02-01 2014-01-28 International Business Machines Corporation Wake-and-go mechanism with dynamic allocation in hardware private array
US8250396B2 (en) 2008-02-01 2012-08-21 International Business Machines Corporation Hardware wake-and-go mechanism for a data processing system
US20090199197A1 (en) * 2008-02-01 2009-08-06 International Business Machines Corporation Wake-and-Go Mechanism with Dynamic Allocation in Hardware Private Array
US20090199030A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Hardware Wake-and-Go Mechanism for a Data Processing System
US8312458B2 (en) 2008-02-01 2012-11-13 International Business Machines Corporation Central repository for wake-and-go mechanism
US8612977B2 (en) 2008-02-01 2013-12-17 International Business Machines Corporation Wake-and-go mechanism with software save of thread state
US8316218B2 (en) 2008-02-01 2012-11-20 International Business Machines Corporation Look-ahead wake-and-go engine with speculative execution
US8341635B2 (en) 2008-02-01 2012-12-25 International Business Machines Corporation Hardware wake-and-go mechanism with look-ahead polling
US8386822B2 (en) 2008-02-01 2013-02-26 International Business Machines Corporation Wake-and-go mechanism with data monitoring
US8452947B2 (en) 2008-02-01 2013-05-28 International Business Machines Corporation Hardware wake-and-go mechanism and content addressable memory with instruction pre-fetch look-ahead to detect programming idioms
US8171476B2 (en) 2008-02-01 2012-05-01 International Business Machines Corporation Wake-and-go mechanism with prioritization of threads
US20090199184A1 (en) * 2008-02-01 2009-08-06 Arimilli Ravi K Wake-and-Go Mechanism With Software Save of Thread State
US8312456B2 (en) * 2008-05-30 2012-11-13 International Business Machines Corporation System and method for optimizing interrupt processing in virtualized environments
US20090300317A1 (en) * 2008-05-30 2009-12-03 International Business Machines Corporation System and method for optimizing interrupt processing in virtualized environments
US20100185847A1 (en) * 2009-01-20 2010-07-22 New York University Database outsourcing with access privacy
US20100268790A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Complex Remote Update Programming Idiom Accelerator
US20100268791A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Programming Idiom Accelerator for Remote Update
US20100269115A1 (en) * 2009-04-16 2010-10-21 International Business Machines Corporation Managing Threads in a Wake-and-Go Engine
US8082315B2 (en) 2009-04-16 2011-12-20 International Business Machines Corporation Programming idiom accelerator for remote update
US8145723B2 (en) 2009-04-16 2012-03-27 International Business Machines Corporation Complex remote update programming idiom accelerator
US8886919B2 (en) 2009-04-16 2014-11-11 International Business Machines Corporation Remote update programming idiom accelerator with allocated processor resources
US8230201B2 (en) 2009-04-16 2012-07-24 International Business Machines Corporation Migrating sleeping and waking threads between wake-and-go mechanisms in a multiple processor data processing system
US20140259014A1 (en) * 2011-10-06 2014-09-11 Hitachi, Ltd. Virtual server processing control method, system, and virtual server processing control management server
US9459898B2 (en) * 2011-10-06 2016-10-04 Hitachi, Ltd. Virtual server processing control method, system, and virtual server processing control management server
US20130232254A1 (en) * 2012-03-02 2013-09-05 Computenext Inc. Cloud resource utilization management
US9513961B1 (en) * 2014-04-02 2016-12-06 Google Inc. Monitoring application loading

Also Published As

Publication number Publication date
US8141083B2 (en) 2012-03-20
US20080184257A1 (en) 2008-07-31

Similar Documents

Publication Publication Date Title
JP5091215B2 (en) System and method for enhancing coprocessor performance
US7577770B2 (en) System and method for performance monitoring and reconfiguring computer system with hardware monitor
CN100559348C (en) Method for providing policy-based operating system services in an operating system and a computing system
US8285966B2 (en) Autonomic self-tuning of database management system in dynamic logical partitioning environment
US10423451B2 (en) Opportunistically scheduling and adjusting time slices
US8631403B2 (en) Method and system for managing tasks by dynamically scaling centralized virtual center in virtual infrastructure
US8806493B2 (en) System and method for providing hardware virtualization in a virtual machine environment
US7174274B2 (en) Gathering I/O measurement data during an I/O operation process
US6941436B2 (en) Method and apparatus for managing memory blocks in a logical partitioned data processing system
EP1600858B1 (en) Method for dynamic optimization of multiplexed resource partitions
US8112757B2 (en) Apparatus and method of dynamically repartitioning a computer system in response to partition workloads
US6871264B2 (en) System and method for dynamic processor core and cache partitioning on large-scale multithreaded, multiprocessor integrated circuits
US9183016B2 (en) Adaptive task scheduling of Hadoop in a virtualized environment
US20110302577A1 (en) Virtual machine migration techniques
US8127086B2 (en) Transparent hypervisor pinning of critical memory areas in a shared memory partition data processing system
US20020087611A1 (en) Virtual computer system with dynamic resource reallocation
US7178145B2 (en) Queues for soft affinity code threads and hard affinity code threads for allocation of processors to execute the threads in a multi-processor system
KR101020392B1 (en) Sharing a kernel of an operating system among logical partitions
JP5433837B2 (en) Virtual computer system, virtual computer control method, and program
US8082400B1 (en) Partitioning a memory pool among plural computing nodes
US8245236B2 (en) Lock based moving of threads in a shared processor partitioning environment
KR100553917B1 (en) Deallocation of computer data in a multithreaded computer
KR20120000066A (en) Virtual non-uniform memory architecture for virtual machines
CN100570565C (en) Method and system for providing policy-based operation system service in management procedure
US7992151B2 (en) Methods and apparatuses for core allocations

Legal Events

Date Code Title Description
AS Assignment

Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:OLSZEWSKI, BRET RONALD;SWANBERG, RANDAL CRAIG;REEL/FRAME:015141/0571;SIGNING DATES FROM 20040901 TO 20040902

Owner name: INTERNATIONAL BUSINESS MACHINES CORPORATION, NEW Y

Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNORS:OLSZEWSKI, BRET RONALD;SWANBERG, RANDAL CRAIG;REEL/FRAME:015141/0554;SIGNING DATES FROM 20040901 TO 20040902

STCB Information on status: application discontinuation

Free format text: EXPRESSLY ABANDONED -- DURING EXAMINATION