Harnessing the Multicores: Nested Data Parallelism in Haskell

Jones, Simon Peyton

doi:10.1007/978-3-540-89330-1_10

Cited by 51 publications

(13 citation statements)

References 24 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For a predecessor of DPH, Nepal, it is shown how a Barnes-Hut algorithm can be implemented in this language [25]. Concrete performance comparisons with this Haskell variant would be interesting future work.…”

Section: Discussionmentioning

confidence: 99%

Parallel Haskell implementations of the N‐body problem

Totoo

Loidl

2013

Concurrency and Computation

View full text Add to dashboard Cite

SUMMARYThis paper provides an assessment of high-level parallel programming models for multi-core programming by implementing two versions of the n-body problem. We compare three different parallel programming models based on parallel Haskell, differing in the ways how potential parallelism is identified and managed. We assess the performance of each implementation, discuss the sequential and parallel tuning steps leading to the final versions, and draw general conclusions on the suitability of high-level parallel programming models for multi-core programming. We achieve speed-ups of up to 7.2 for the all-pairs algorithm and up to 6.5 for the Barnes-Hut algorithm on an 8-core machine.

show abstract

Section: Discussionmentioning

confidence: 99%

Parallel Haskell implementations of the N‐body problem

Totoo

Loidl

2013

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…We focus specifically on improving co-scheduling between jobs using abstractions such as CDDP [12], Cilk [11], Data-Parallel Haskell [30], Galois [17], and OpenMP. In these settings, workloads identify sources of parallelism without explicitly forking threads.…”

Section: Setting and Assumptionsmentioning

confidence: 99%

Callisto

Harris

Maas

Marathe

2014

Proceedings of the Ninth European Conference on Computer Systems

View full text Add to dashboard Cite

It is increasingly important for parallel applications to run together on the same machine. However, current performance is often poor: programs do not adapt well to dynamically varying numbers of cores, and the CPU time received by concurrent jobs can differ drastically. This paper introduces Callisto, a resource management layer for parallel runtime systems. We describe Callisto and the implementation of two Callisto-enabled runtime systems-one for OpenMP, and another for a task-parallel programming model. We show how Callisto eliminates almost all of the scheduler-related interference between concurrent jobs, while still allowing jobs to claim otherwise-idle cores. We use examples from two recent graph analytics projects and from SPEC OMP.

show abstract

“…Pure dataparallel languages follow this strategy [28], as do languages that force references to be either task-unique or immutable [5]. But some algorithms are more naturally or efficiently written using shared state or message passing.…”

Section: Introductionmentioning

confidence: 99%

Freeze after writing

Kuper

Turon

Krishnaswami

et al. 2014

Proceedings of the 41st ACM SIGPLAN-SIGACT Symposium on Principles of Programming Languages

View full text Add to dashboard Cite

Deterministic-by-construction parallel programming models offer the advantages of parallel speedup while avoiding the nondeterministic, hard-to-reproduce bugs that plague fully concurrent code. A principled approach to deterministic-by-construction parallel programming with shared state is offered by LVars: shared memory locations whose semantics are defined in terms of an applicationspecific lattice. Writes to an LVar take the least upper bound of the old and new values with respect to the lattice, while reads from an LVar can observe only that its contents have crossed a specified threshold in the lattice. Although it guarantees determinism, this interface is quite limited.We extend LVars in two ways. First, we add the ability to "freeze" and then read the contents of an LVar directly. Second, we add the ability to attach event handlers to an LVar, triggering a callback when the LVar's value changes. Together, handlers and freezing enable an expressive and useful style of parallel programming. We prove that in a language where communication takes place through these extended LVars, programs are at worst quasideterministic: on every run, they either produce the same answer or raise an error. We demonstrate the viability of our approach by implementing a library for Haskell supporting a variety of LVarbased data structures, together with a case study that illustrates the programming model and yields promising parallel speedup.

show abstract

Harnessing the Multicores: Nested Data Parallelism in Haskell

Cited by 51 publications

References 24 publications

Parallel Haskell implementations of the N‐body problem

Parallel Haskell implementations of the N‐body problem

Callisto

Freeze after writing

Contact Info

Product

Resources

About