Adaptive Implementation Selection in the SkePU Skeleton Programming Library

Dastgeer, Usman; Li, Lü; Keßler, Christoph

doi:10.1007/978-3-642-45293-2_13

Cited by 22 publications

(33 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We therefore exclude adi, cholesky, dynprog, durbin, fdtd2d-apml, gramschmidt, lu, ludcmp, reg_detect, symm, trmm, and trisolv. This exclusion can be automated by integrating a performance model (e.g., Dastgeer et al [2013]). All in all, 34 (not necessarily unique) species are included, spread across 16 benchmarks.…”

Section: Resultsmentioning

confidence: 99%

See 1 more Smart Citation

Bones

Nugteren

Corporaal

2014

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

The shift toward parallel processor architectures has made programming and code generation increasingly challenging. To address this programmability challenge, this article presents a technique to fully automatically generate efficient and readable code for parallel processors (with a focus on GPUs). This is made possible by combining algorithmic skeletons, traditional compilation, and "algorithmic species," a classification of program code. Compilation starts by automatically annotating C code with class information (the algorithmic species). This code is then fed into the skeleton-based source-to-source compiler BONES to generate CUDA code. To generate efficient code, BONES also performs optimizations including host-accelerator transfer optimization and kernel fusion. This results in a unique approach, integrating a skeleton-based compiler for the first time into an automated flow. The benefits are demonstrated experimentally for PolyBench GPU kernels, showing geometric mean speed-ups of 1.4× and 2.4× compared to PPCG and PAR4ALL, and for five Rodinia GPU benchmarks, showing a gap of only 1.2× compared to hand-optimized code.

show abstract

Section: Resultsmentioning

confidence: 99%

“…The compiler first loads and invokes the corresponding skeletons for a given target. The target is specified by the user but could be selected dynamically based on the invoked skeleton and data sizes, as is done by SkePU [Dastgeer et al 2013]. Next, Fig.…”

Section: A Skeleton-based Source-to-source Compilermentioning

confidence: 99%

Bones

Nugteren

Corporaal

2014

ACM Trans. Archit. Code Optim.

View full text Add to dashboard Cite

show abstract

“…This scheduler chooses the most appropriate task version among those marked as implementation alternatives. Another solution is the extension for the SkePu framework [6], which leverages machine learning techniques to decide which of the available versions of a given function offers the lowest execution time. Following a similar approach, the selector presented in this paper uses a novel technique based on probabilities and problem sizes that allows determining the best implementation at runtime.…”

Section: Related Workmentioning

confidence: 99%

Probabilistic-Based Selection of Alternate Implementations for Heterogeneous Platforms

Fernández¹,

Cuadrado²,

Astorga³

et al. 2017

Algorithms and Architectures for Parallel Processing

View full text Add to dashboard Cite

Abstract. Over the last years, heterogeneous architectures have become a de facto approach for improving the performance of numerous scientific and industrial applications. However, developing for these architectures is not straightforward: each processor demands its specific programming paradigm and, often, certain applications are only well-suited to run on a particular processing unit. Therefore, a major challenge arises when programming for these platforms: to select the most suitable device and routine implementation to solve a given problem. To deal with this issue, this paper proposes a novel probabilistic-based selector that uses the problem size to automatically choose the most appropriate version of a same kernel. In order to analyze this approach, we have developed this selector within the OmpSs programming framework and evaluated its accuracy and performance gains when executing different implementations of the general matrix-matrix multiplication. Finally, we also demonstrate how this solution delivers a comparable performance with respect to a runtime approach from the state-of-the-art.

show abstract

“…We refer to earlier SkePU publications [5,7,9] for other work relating to specific features, such as smart containers.…”

Section: Related Workmentioning

confidence: 99%

“…SkePU [9] is a C++ template library for high-level performance-portable parallel programming, based on algorithmic skeletons [3]. SkePU has previously been demonstrated to provide an approachable interface [9], an auto-tuning mechanism for backend selection [7] and efficient smart containers [5]. Several industrial-class applications have been ported to SkePU, for example the computational fluid dynamics flow solver EDGE [20].…”

Section: Skepumentioning

confidence: 99%

SkePU 2: Flexible and Type-Safe Skeleton Programming for Heterogeneous Parallel Systems

Ernstsson

Keßler

2017

Int J Parallel Prog

Self Cite

View full text Add to dashboard Cite

In this article we present SkePU 2, the next generation of the SkePU C++ skeleton programming framework for heterogeneous parallel systems. We critically examine the design and limitations of the SkePU 1 programming interface. We present a new, flexible and type-safe, interface for skeleton programming in SkePU 2, and a source-to-source transformation tool which knows about SkePU 2 constructs such as skeletons and user functions. We demonstrate how the source-to-source compiler transforms programs to enable efficient execution on parallel heterogeneous systems. We show how SkePU 2 enables new use-cases and applications by increasing the flexibility from SkePU 1, and how programming errors can be caught earlier and easier thanks to improved type safety. We propose a new skeleton, Call, unique in the sense that it does not impose any predefined skeleton structure and can encapsulate arbitrary user-defined multi-backend computations. We also discuss how the sourceto-source compiler can enable a new optimization opportunity by selecting among multiple user function specializations when building a parallel program. Finally, we show that the performance of our prototype SkePU 2 implementation closely matches that of SkePU 1.

show abstract

Adaptive Implementation Selection in the SkePU Skeleton Programming Library

Cited by 22 publications

References 7 publications

Bones

Bones

Probabilistic-Based Selection of Alternate Implementations for Heterogeneous Platforms

SkePU 2: Flexible and Type-Safe Skeleton Programming for Heterogeneous Parallel Systems

Contact Info

Product

Resources

About