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,

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

In this paper we prove time and space bounds for the implementation of the programming language NESL on various parallel machine models. NESL is a sugared typed J-calculus with a set of array primitives and an explicit parallel map over arrays. Our results extend previous work on provable implementation bounds for functional languages by considering space and by including arrays. For modeling the cost of NESL we augment a standard call-by-value operational semantics to return two cost measures: a DAG representing the sequential dependence in the computation, and a measure of the space taken by a sequential implementation. We show that a NESL program with w work (nodes in the DAG), d depth (levels in the DAG), and s sequential space can be implemented on a p processor butterfly network, hypercube, or CRCW PRAM usin O(w/p + d log p) time and 0(s + dp logp) reachable space. For programs with sufficient parallelism these bounds are optimal in that they give linew speedup and use space within a constant factor of the sequential space. 1

Abstract not found

Recent work on scheduling algorithms has resulted in provable bounds on the space taken by parallel computations in relation to the space taken by sequential computations. The results for online versions of these algorithms, however, have been limited to computations in which threads can only synchronize with ancestor or sibling threads. Such computations do not include languages with futures or user-specified synchronization constraints. Here we extend the results to languages with synchronization variables. Such languages include languages with futures, such as Multilisp and Cool, as well as other languages such asid. The main result is an online scheduling algorithm which, given a computation with w work (total operations), synchronizations, d depth (critical path) and s1 sequential space, will run in O(w=p + log(pd)=p + d log(pd)) time and s1 + O(pd log(pd)) space, on a p-processor crcw pram with a fetch-and-add primitive. This includes all time and space costs for both the computation and the scheduler. The scheduler is non-preemptive in the sense that it will only move a thread if the thread suspends on a synchronization, forks a new thread, or exceeds a threshold when allocating space. For the special case where the computation is a planar graph with left-to-right synchronization edges, the scheduling algorithm can be implemented in O(w=p+d log p) time and s1 + O(pd log p) space. These are the first nontrivial space bounds described for such languages.

This article presents an on-line scheduling algorithm that is provably space e#cient and time e#cient for nested-parallel languages. For a computation with depth D and serial space requirement S1 , the algorithm generates a schedule that requires at most S1 +O(K D p)space (including scheduler space) on p processors. Here, K is a user-adjustable runtime parameter specifying the net amount of memory that a thread may allocate before it is preempted by the scheduler. Adjusting the value of K provides a trade-o# between the running time and the memory requirement of a parallel computation. To allow the scheduler to scale with the number of processors, we also parallelize the scheduler and analyze the space and time bounds of the computation to include scheduling costs. In addition to showing that the scheduling algorithm is space and time e#cient in theory, we demonstrate that it is e#ective in practice. We have implemented a runtime system that uses our algorithm to schedule lightweight parallel threads. The results of executing parallel programs on this system show that our scheduling algorithm significantly reduces memory usage compared to previous techniques, without compromising performance

Abstract not found

This paper presents a multicore-cache model that reflects the reality that multicore processors have both per-processor private (L1) caches and a large shared (L2) cache on chip. We consider a broad class of parallel divide-andconquer algorithms and present a new on-line scheduler, controlled-pdf, that is competitive with the standard sequential scheduler in the following sense. Given any dynamically unfolding computation DAG from this class of algorithms, the cache complexity on the multicore-cache model under our new scheduler is within a constant factor of the sequential cache complexity for both L1 and L2, while the time complexity is within a constant factor of the sequential time complexity divided by the number of processors p. These are the first such asymptoticallyoptimal results for any multicore model. Finally, we show that a separator-based algorithm for sparse-matrix-densevector-multiply achieves provably good cache performance in the multicore-cache model, as well as in the well-studied sequential cache-oblivious model.

We compare the number of cache misses M1 for running a computation on a single processor with cache size C1 to the total number of misses Mp for the same computation when using p processors or threads and a shared cache of size Cp . We show that for any computation, and with an appropriate (greedy) parallel schedule, if Cp C1 + pd then Mp M1 . The depth d of the computation is the length of the critical path of dependences. This gives the perhaps surprising result that for sufficiently parallel computations the shared cache need only be an additive size larger than the singleprocessor cache, and gives some theoretical justification for designing machines with shared caches. We model

The running time and memory requirement of a parallel program with dynamic, lightweight threads depends heavily on the underlying thread scheduler. In this paper, we present a simple, asynchronous, space-efficient scheduling algorithm for shared memory machines that combines the low scheduling overheads and good locality of work stealing with the low space requirements of depth-first schedulers. For a nested-parallel program with depth D and serial space requirement S 1 , we show that the expected space requirement is S 1 +O(K \Delta p \Delta D) on p processors. Here, K is a user-adjustable runtime parameter, which provides a tradeoff between running time and space requirement. Our algorithm achieves good locality and low scheduling overheads by automatically increasing the granularity of the work scheduled on each processor. We have implemented the new scheduling algorithm in the context of a native, user-level implementation of Posix standard threads or Pthreads, and evaluated its p...