International Business Machines Corporation

Lawrence Livermore National Security, LLC

Larry Bert Brenner

Peter Richard Badovinatz

Orvalle Theodore Kirby

John Arthur Pershing, Jr

Tushar Deepak Chandra

Luke Matthew Browning

Marc Adam Kaplan

Ajei Sarat Gopal

Marion Lee Blount

John Joseph Turek

Kenneth C Briskey

Krystal Kay Rothaupt

Dennis Duane Jurgensen

Mysore Sathyanarayana Srinivas

James William Van Fleet

Luke Mathew Browning

John Joseph Edward Turek

Barry P Lubart

Bruce Gerard Mealey

Bret Ronald Olszewski

Dirk Michel

Basu Vaidyanathan

Dean J Burdick

Jeffrey S Lucash

Greg Robert Mewhinney

James Bernard Moody

Michael E Lyons

Patrick Francis Caffrey

Kenneth Bernard Rozendal

James William VanFleet

John C Rathjen, Jr

Angelo Pruscino

Thomas Van Weaver

Paul W VanLeer

Bimal Kiran Doshi

Ronald Sasala

Terry R Jones

Patrick O'Rourke

Pythagoras C Watson

Jeffrey R Hedglin

William Tuel

Jeffrey Fier

The illustrative embodiments provide a computer implemented method, apparatus, and computer usable program code for managing a heap. The heap is partitioned into at least one sub heap based on a relationship to at least one memory class of a plurality of memory classes. A memory allocation request comprising a memory class is received from a requester. A unique heap handle based on the memory class and associated with a specific sub heap is generated. The unique heap handle is then returned to the requester.

Memory class based heap partitioning

The present invention provides a computer implemented method and apparatus to assign software threads to a common virtual processor of a data processing system having multiple virtual processors. A data processing system detects cooperation between a first thread and a second thread with respect to a lock associated with a resource of the data processing system. Responsive to detecting cooperation, the data processing system assigns the first thread to the common virtual processor. The data processing system moves the second thread to the common virtual processor, whereby a sleep time associated with the lock experienced by the first thread and the second thread is reduced below a sleep time experienced prior to the detecting cooperation step.

Lock based moving of threads in a shared processor partitioning environment

A computer implemented method, a data processing system, and computer usable program code for improving thread posting efficiency in a multiprocessor data processing system are provided. Aspects of the present invention first receive a set of threads from an application. The aspects of the present invention then group the set of threads with a plurality of processors based on a last execution of the set of threads on the plurality of processors to form a plurality of groups. The threads in each group in the plurality of groups are all last executed on a same processor. The aspects of the present invention then wake up the threads in the plurality of groups in any order.

Method and apparatus for awakening client threads in a multiprocessor data processing system

In a parallel computing environment comprising a network of SMP nodes each having at least one processor, a parallel-aware co-scheduling method and system for improving the performance and scalability of a dedicated parallel job having synchronizing collective operations. The method and system uses a global co-scheduler and an operating system kernel dispatcher adapted to coordinate interfering system and daemon activities on a node and across nodes to promote intra-node and inter-node overlap of said interfering system and daemon activities as well as intra-node and inter-node overlap of said synchronizing collective operations. In this manner, the impact of random short-lived interruptions, such as timer-decrement processing and periodic daemon activity, on synchronizing collective operations is minimized on large processor-count SPMD bulk-synchronous programming styles.

Parallel-aware, dedicated job co-scheduling within/across symmetric multiprocessing nodes

A system, apparatus and method of enhancing priority boosting of scheduled threads are provided. If, while being executed by a second CPU, a second thread determines that it has to wait for a lock on a shared resource held by a first thread that is scheduled to be executed by a first CPU, the second thread may boost the priority of the first thread by passing its priority to the first thread if its priority is higher than the first thread's priority. Further, to enhance the priority boost of the first thread, the second thread may reschedule the first thread to be processed by the second CPU. By having been rescheduled on the second CPU, the second thread may be dispatched for execution right thereafter.

System, apparatus and method of enhancing priority boosting of scheduled threads

An operating system in a shared processor logical partitioned data processing system is given a target percentage. The hypervisor assigns the target processor percentage to the operating system. The operating system also has a predetermined time slice to allot to threads in a multitasking environment. The operating system adjusts the time slice based on a per-virtual-processor percentage.

Time slicing in a shared partition

A method, system, and computer instructions for atomically updating 64-bit fields in the 32-bit AIX kernel, wherein the underlying hardware's 64-bit capable instructions are used to construct an atomic primitive. If a data request applies to the 32-bit kernel on 64-bit hardware execution environment, a load and reserve instruction sets a reservation for a memory location. The original 64-bit data object is decomposed into two 32-bit halves. A Shift Right Double Word Immediate (SRDI) instruction captures the high-order bits of the 64-bit register. If the store conditional instruction determines that the reservation is not lost, the store conditional instruction stores the result. If the store conditional instruction fails, the process returns to the reserve instruction until the store conditional operation returns a success.

Atomically updating 64 bit fields in the 32 bit AIX kernel

A system and method for thread scheduling with a weak preemption policy is provided. The scheduler receives requests from newly ready work. The scheduler adds a 'preempt value' to the current work's priority so that it is somewhat increased for preemption purposes. The preempt value can be adjusted in order to make it more, or less, difficult for newly ready work to preempt the current work. A 'less strict' preemption policy allows current work to complete rather than interrupting the current work and resume it at a later time, thus saving system overhead. Newly ready work that is queued with a better priority than the current work is queued in a favorable position to be executed after the current work is completed but before other work that has been queued with the same priority of the current work.

System and method for thread scheduling with weak preemption policy

A system and method is provided for using a kernel exit routine, performed when a parent process exits, to efficiently remove zombie child processes. The kernel exit routine also re-parents active child processes to a standard system process, such as the init process. When the kernel exit routine re-parents the active child process to the init process, the child process is flagged indicating that the child's original parent process has already exited. A kernel routine, such as the swapper process, periodically cleans up flagged processes that have exited. In this manner, although active child processes are still re-parented to init, the init process is no longer burdened with the task of waiting on such processes when they exit.

System and method to improve harvesting of zombie processes in an operating system

An application implementing a high resolution monotonic system clock is provided. A kernel increments a value called a generation number. The generation number is used to determine what state the high and low resolution time values are in. The kernel next stores a low resolution time value. Then, if a time adjustment is occurring, the kernel updates the high resolution time value by updating the origin date. The kernel then increments the generation number a second time. The providing application receives a request for time from another application. The providing application determines a first value for the generation number and compares a low resolution time value and high resolution time value and determines the maximum value of the two. The providing application then gets a second value for the generation number and compares the second value of the generation number to the first value of the generation number. If the two values of the generation number are the same, the providing application returns the maximum value to the requesting application.

Implementing a high resolution monotonic system clock

Method and system for detecting and reporting an excessive period of interrupt disablement in operating system programming in a data processing system. A method for detecting and reporting an excessive period of disablement in operating system programming in a data processing system includes determining, at a plurality of scheduled times, if an operating system task is running disabled for interrupts, and reporting an excessive disablement if the operating system task runs disabled for interrupts for a predefined number of consecutive scheduled times.

Automatic detection of excessive interrupt-disabled operating system code

A method, system and apparatus for integrating a system task scheduler with a workload manager are provided. The scheduler is used to assign default priorities to threads and to place the threads into run queues and the workload manager is used to implement policies set by a system administrator. One of the policies may be to have different classes of threads get different percentages of a system's CPU time. This policy can be reliably achieved if threads from a plurality of classes are spread as uniformly as possible among the run queues. To do so, the threads are organized in classes. Each class is associated with a priority as per a use-policy. This priority is used to modify the scheduling priority assigned to each thread in the class as well as to determine in which band or range of priority the threads fall. Then periodically, it is determined whether the number of threads in a band in a run queue exceeds the number of threads in the band in another run queue by more than a pre-determined number. If so, the system is deemed to be load-imbalanced. If not, the system is load-balanced by moving one thread in the band from the run queue with the greater number of threads to the run queue with the lower number of threads.

Multiprocessor load balancing system for prioritizing threads and assigning threads into one of a plurality of run queues based on a priority band and a current load of the run queue

Apparatus and methods for load balancing fixed priority threads in a multiprocessor system are provided. The apparatus and methods of the present invention identify unbound fixed priority threads at the top of local run queues. A best fixed priority thread is then identified and its priority checked against the priorities of threads executing on processors of the node. A set of executing threads that may be displaced by the best fixed priority thread is identified. The lowest priority executing thread from the set is then identified and the best fixed priority thread is moved to displace this lowest priority executing thread.

Apparatus and method for load balancing of fixed priority threads in a multiple run queue environment

Growing community of inventors

Austin, TX, United States of America

Larry Bert Brenner