Results 1 - 10
of
487
Scheduling Multithreaded Computations by Work Stealing
, 1994
"... This paper studies the problem of efficiently scheduling fully strict (i.e., well-structured) multithreaded computations on parallel computers. A popular and practical method of scheduling this kind of dynamic MIMD-style computation is “work stealing," in which processors needing work steal com ..."
Abstract
-
Cited by 572 (43 self)
- Add to MetaCart
This paper studies the problem of efficiently scheduling fully strict (i.e., well-structured) multithreaded computations on parallel computers. A popular and practical method of scheduling this kind of dynamic MIMD-style computation is “work stealing," in which processors needing work steal computational threads from other processors. In this paper, we give the first provably good work-stealing scheduler for multithreaded computations with dependencies. Specifically, our analysis shows that the ezpected time Tp to execute a fully strict computation on P processors using our work-stealing scheduler is Tp = O(TI/P + Tm), where TI is the minimum serial eze-cution time of the multithreaded computation and T, is the minimum ezecution time with an infinite number of processors. Moreover, the space Sp required by the execution satisfies Sp 5 SIP. We also show that the ezpected total communication of the algorithm is at most O(TmS,,,P), where S, is the site of the largest activation record of any thread, thereby justify-ing the folk wisdom that work-stealing schedulers are more communication eficient than their work-sharing counterparts. All three of these bounds are existentially optimal to within a constant factor.
Thread scheduling for multiprogrammed multiprocessors
- In Proceedings of the Tenth Annual ACM Symposium on Parallel Algorithms and Architectures (SPAA), Puerto Vallarta
, 1998
"... We present a user-level thread scheduler for shared-memory multiprocessors, and we analyze its performance under multiprogramming. We model multiprogramming with two scheduling levels: our scheduler runs at user-level and schedules threads onto a fixed collection of processes, while below, the opera ..."
Abstract
-
Cited by 213 (5 self)
- Add to MetaCart
(Show Context)
We present a user-level thread scheduler for shared-memory multiprocessors, and we analyze its performance under multiprogramming. We model multiprogramming with two scheduling levels: our scheduler runs at user-level and schedules threads onto a fixed collection of processes, while below, the operating system kernel schedules processes onto a fixed collection of processors. We consider the kernel to be an adversary, and our goal is to schedule threads onto processes such that we make efficient use of whatever processor resources are provided by the kernel. Our thread scheduler is a non-blocking implementation of the work-stealing algorithm. For any multithreaded computation with work ¢¤ £ and critical-path length ¢¦ ¥ , and for any number § of processes, our scheduler executes the computation in expected time ¨�©�¢�£���§¤����¢�¥�§���§¤�� � , where §� � is the average number of processors allocated to the computation by the kernel. This time bound is optimal to within a constant factor, and achieves linear speedup whenever § is small relative to the parallelism 1
A Fast Fourier Transform Compiler
, 1999
"... FFTW library for computing the discrete Fourier transform (DFT) has gained a wide acceptance in both academia and industry, because it provides excellent performance on a variety of machines (even competitive with or faster than equivalent libraries supplied by vendors). In FFTW, most of the perform ..."
Abstract
-
Cited by 193 (5 self)
- Add to MetaCart
(Show Context)
FFTW library for computing the discrete Fourier transform (DFT) has gained a wide acceptance in both academia and industry, because it provides excellent performance on a variety of machines (even competitive with or faster than equivalent libraries supplied by vendors). In FFTW, most of the performance-critical code was generated automatically by a special-purpose compiler, called genfft, that outputs C code. Written in Objective Caml, genfft can produce DFT programs for any input length, and it can specialize the DFT program for the common case where the input data are real instead of complex. Unexpectedly, genfft “discovered” algorithms that were previously unknown, and it was able to reduce the arithmetic complexity of some other existing algorithms. This paper describes the internals of this special-purpose compiler in some detail, and it argues that a specialized compiler is a valuable tool.
Iterative Context Bounding for Systematic Testing of Multithreaded Programs
, 2007
"... Multithreaded programs are difficult to get right because of unexpected interaction between concurrently executing threads. Traditional testing methods are inadequate for catching subtle concurrency errors which manifest themselves late in the development cycle and post-deployment. Model checking or ..."
Abstract
-
Cited by 177 (16 self)
- Add to MetaCart
Multithreaded programs are difficult to get right because of unexpected interaction between concurrently executing threads. Traditional testing methods are inadequate for catching subtle concurrency errors which manifest themselves late in the development cycle and post-deployment. Model checking or systematic exploration of program behavior is a promising alternative to traditional testing methods. However, it is difficult to perform systematic search on large programs as the number of possible program behaviors grows exponentially with the program size. Confronted with this state-explosion problem, traditional model checkers perform iterative depth-bounded search. Although effective for message-passing software, iterative depth-bounding is inadequate for multithreaded software. This paper proposes iterative context-bounding, a new search algorithm that systematically explores the executions of a multithreaded program in an order that prioritizes executions with fewer context switches. We distinguish between preempting and nonpreempting context switches, and show that bounding the number of preempting context switches to a small number significantly alleviates the state explosion, without limiting the depth of explored executions. We show both theoretically and empirically that contextbounded search is an effective method for exploring the behaviors of multithreaded programs. We have implemented our algorithm in two model checkers and applied it to a number of real-world multithreaded programs. Our implementation uncovered 9 previously unknown bugs in our benchmarks, each of which was exposed by an execution with at most 2 preempting context switches. Our initial experience with the technique is encouraging and demonstrates that iterative context-bounding is a significant improvement over existing techniques for testing multithreaded programs.
Pointer Analysis for Multithreaded Programs
- ACM SIGPLAN 99
, 1999
"... This paper presents a novel interprocedural, flow-sensitive, and context-sensitive pointer analysis algorithm for multithreaded programs that may concurrently update shared pointers. For each pointer and each program point, the algorithm computes a conservative approximation of the memory locations ..."
Abstract
-
Cited by 158 (11 self)
- Add to MetaCart
This paper presents a novel interprocedural, flow-sensitive, and context-sensitive pointer analysis algorithm for multithreaded programs that may concurrently update shared pointers. For each pointer and each program point, the algorithm computes a conservative approximation of the memory locations to which that pointer may point. The algorithm correctly handles a full range of constructs in multithreaded programs, including recursive functions, function pointers, structures, arrays, nested structures and arrays, pointer arithmetic, casts between pointer variables of different types, heap and stack allocated memory, shared global variables, and thread-private global variables. We have implemented the algorithm in the SUIF compiler system and used the implementation to analyze a sizable set of multithreaded programs written in the Cilk multithreaded programming language. Our experimental results show that the analysis has good precision and converges quickly for our set of Cilk programs.
Symbolic Bounds Analysis of Pointers, Array Indices, and Accessed Memory Regions
- PLDI 2000
, 2000
"... This paper presents a novel framework for the symbolic bounds analysis of pointers, array indices, and accessed memory regions. Our framework formulates each analysis problem as a system of inequality constraints between symbolic bound polynomials. It then reduces the constraint system to a linear p ..."
Abstract
-
Cited by 135 (15 self)
- Add to MetaCart
(Show Context)
This paper presents a novel framework for the symbolic bounds analysis of pointers, array indices, and accessed memory regions. Our framework formulates each analysis problem as a system of inequality constraints between symbolic bound polynomials. It then reduces the constraint system to a linear program. The solution to the linear program provides symbolic lower and upper bounds for the values of pointer and array index variables and for the regions of memory that each statement and procedure accesses. This approach eliminates fundamental problems associated with applying standard xed-point approaches to symbolic analysis problems. Experimental results from our implemented compiler show that the analysis can solve several important problems, including static race detection, automatic parallelization, static detection of array bounds violations, elimination of array bounds checks, and reduction of the number of bits used to store computed values.
A Java Fork/Join Framework
, 2000
"... This paper describes the design, implementation, and performance of a Java framework for supporting a style of parallel programming in which problems are solved by (recursively) splitting them into subtasks that are solved in parallel, waiting for them to complete, and then composing results. The ge ..."
Abstract
-
Cited by 112 (0 self)
- Add to MetaCart
This paper describes the design, implementation, and performance of a Java framework for supporting a style of parallel programming in which problems are solved by (recursively) splitting them into subtasks that are solved in parallel, waiting for them to complete, and then composing results. The general design is a variant of the work-stealing framework devised for Cilk. The main implementation techniques surround efficient construction and management of tasks queues and worker threads. The measured performance shows good parallel speedups for most programs, but also suggests possible improvements. 1. INTRODUCTION Fork/Join parallelism is among the simplest and most effective design techniques for obtaining good parallel performance. Fork/join algorithms are parallel versions of familiar divideand -conquer algorithms, taking the typical form: Result solve(Problem problem) { if (problem is small) directly solve problem else { split problem into independent parts fork new subtas...
Kendo: Efficient Deterministic Multithreading in Software
- In ASPLOS
, 2009
"... Although chip-multiprocessors have become the industry standard, developing parallel applications that target them remains a daunting task. Non-determinism, inherent in threaded applications, causes significant challenges for par-allel programmers by hindering their ability to create parallel applic ..."
Abstract
-
Cited by 109 (0 self)
- Add to MetaCart
(Show Context)
Although chip-multiprocessors have become the industry standard, developing parallel applications that target them remains a daunting task. Non-determinism, inherent in threaded applications, causes significant challenges for par-allel programmers by hindering their ability to create parallel applications with repeatable results. As a consequence, par-allel applications are significantly harder to debug, test, and maintain than sequential programs. This paper introduces Kendo: a new software-only sys-tem that provides deterministic multithreading of parallel applications. Kendo enforces a deterministic interleaving of lock acquisitions and specially declared non-protected reads through a novel dynamically load-balanced deterministic scheduling algorithm. The algorithm tracks the progress of each thread using performance counters to construct a deterministic logical time that is used to compute an inter-leaving of shared data accesses that is both deterministic and provides good load balancing. Kendo can run on to-day’s commodity hardware while incurring only a modest performance cost. Experimental results on the SPLASH-2 applications yield a geometric mean overhead of only 16% when running on 4 processors. This low overhead makes it possible to benefit from Kendo even after an application is deployed. Programmers can start using Kendo today to pro-gram parallel applications that are easier to develop, debug, and test.
Detecting Data Races in Cilk Programs that Use Locks
, 1998
"... When two parallel threads holding no locks in common access the same memory location and at least one of the threads modifies the location, a “data race ” occurs, which is usually a bug. This paper describes the algorithms and strategies used by a debugging tool, called the Nondeterminator-2, which ..."
Abstract
-
Cited by 90 (11 self)
- Add to MetaCart
(Show Context)
When two parallel threads holding no locks in common access the same memory location and at least one of the threads modifies the location, a “data race ” occurs, which is usually a bug. This paper describes the algorithms and strategies used by a debugging tool, called the Nondeterminator-2, which checks for data races in pro-grams coded in the Cilk multithreaded language. Like its predeces-sor, the Nondeterminator, which checks for simple “determinacy” races, the Nondeterminator-2 is a debugging tool, not a verifier, since it checks for data races only in the computation generated by a serial execution of the program on a given input. We give an algorithm, ALL-SETS, that determines whether the computation generated by a serial execution of a Cilk program on a given input contains a race. For a program that runs serially in time T, accesses V shared memory locations, uses a total of n locks, and holds at most k << n locks simultaneously, ALL-SETS runs in
