Hybridizing S3D into an Exascale application using OpenACC: An approach for moving to multi-petaflops and beyond

Levesque, John; Sankaran, Ramanan; Grout, Ray

doi:10.1109/sc.2012.69

Cited by 36 publications

(26 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Thanks to the support from numerous vendors, OpenACC and OpenMP have rapidly established themselves as the de facto solutions for directive-based code development. Although capable of delivering acceptable performance [44,19] in a broad range of applications, neither OpenACC nor OpenMP targets an entire cluster. Users are thus left to their own to write code that deals with MPI.…”

Section: Compiler Directivesmentioning

confidence: 99%

Panda: A Compiler Framework for Concurrent CPU $$+$$ + GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

Sourouri

Baden

Cai

2016

Int J Parallel Prog

View full text Add to dashboard Cite

This paper describes a new compiler framework for heterogeneous 3D stencil computation on GPU clusters. Our framework consists of a simple directive-based programming model and a tightly integrated source-to-source compiler. Annotated with a small number of directives, sequential stencil codes originally written in C can be automatically parallelized for large-scale GPU clusters. The most distinctive feature of the compiler is its capability to generate state-of-the-art hybrid MPI+CUDA+OpenMP code that uses concurrent CPU+GPU computing to unleash the full potential of powerful GPU clusters. At the same time, the auto-generated hybrid codes hide the overhead of various data motion by overlapping them with computation. Test results on the Titan supercomputer and the Wilkes cluster show that auto-translated codes from our compiler can achieve about 90% of the performance of highly optimized handwritten codes, for both a simple stencil benchmark and a real-world application in cardiac modeling. We thus believe that the user-friendliness and performance delivered by our domain-specific compiler framework allow com-

show abstract

Section: Compiler Directivesmentioning

confidence: 99%

Panda: A Compiler Framework for Concurrent CPU $$+$$ + GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

Sourouri

Baden

Cai

2016

Int J Parallel Prog

View full text Add to dashboard Cite

show abstract

“…Hart et al [8] used Co-Array Fortran (CAF) + OpenACC and Levesque et al [10] used MPI + OpenACC to utilize multi-GPU in GPU cluster. The inter-GPU communication in these cases are managed using approach similar to distributed shared memory model.…”

Section: Related Workmentioning

confidence: 99%

Supporting multiple accelerators in high-level programming models

Lin

Liao

Supinski

et al. 2015

Proceedings of the Sixth International Workshop on Programming Models and Applications for Multicores and Manycores

View full text Add to dashboard Cite

Computational accelerators, such as manycore NVIDIA GPUs, Intel Xeon Phi and FPGAs, are becoming common in workstations, servers and supercomputers for scientific and engineering applications. Efficiently exploiting the massive parallelism these accelerators provide requires the designs and implementations of productive programming models.In this paper, we explore support of multiple accelerators in high-level programming models. We design novel language extensions to OpenMP to support offloading data and computation regions to multiple accelerators (devices). These extensions allow for distributing data and computation among a list of devices via easy-to-use interfaces, including specifying the distribution of multi-dimensional arrays and declaring shared data regions among accelerators. Computation distribution is realized by partitioning a loop iteration space among accelerators. We implement mechanisms to marshal/unmarshal and to move data of non-contiguous array subregions and shared regions between accelerators without involving CPUs. We design reduction techniques that work across multiple accelerators. Combined compiler and runtime support is designed to manage multiple GPUs using asynchronous operations and threading mechanisms. We implement our solutions for NVIDIA GPUs and demonstrate through example OpenMP codes the effectiveness of our solutions for performance improvement.

show abstract

“…Since these are a small subset of the total fields in a Cell, there is a significant improvement in performance by only moving the needed field data. We note that a version of S3D using OpenACC [19] can perform a similar operation, under the condition that data is laid out in system memory using a struct-of-arrays format so individual field data is dense, allowing OpenACC to copy individual dimensions of the array. Entangling layout with data movement optimizations in the OpenACC code results in code that is difficult to modify when exploring different mapping strategies and tuning for new architectures.…”

Section: Motivationmentioning

confidence: 99%

“…We compare against two versions of S3D: a CPU-only version and an improved hybrid version from [19] that uses OpenACC.…”

Section: S3dmentioning

confidence: 99%

See 1 more Smart Citation

Structure Slicing: Extending Logical Regions with Fields

Bauer

Treichler

Slaughter

et al. 2014

SC14: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Abstract-Applications on modern supercomputers are increasingly limited by the cost of data movement, but mainstream programming systems have few abstractions for describing the structure of a program's data. Consequently, the burden of managing data movement, placement, and layout currently falls primarily upon the programmer.To address this problem we previously proposed a data model based on logical regions and described Legion, a programming system incorporating logical regions. In this paper, we present structure slicing, which incorporates fields into the logical region data model. We show that structure slicing enables Legion to automatically infer task parallelism from field non-interference, decouple the specification of data usage from layout, and reduce the overall amount of data moved. We demonstrate that structure slicing enables both strong and weak scaling of three Legion applications including S3D, a production combustion simulation that uses logical regions with thousands of fields, with speedups of up to 3.68X over a vectorized CPU-only Fortran implementation and 1.88X over an independently hand-tuned OpenACC code.

show abstract

Hybridizing S3D into an Exascale application using OpenACC: An approach for moving to multi-petaflops and beyond

Cited by 36 publications

References 5 publications

Panda: A Compiler Framework for Concurrent CPU $$+$$ + GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

Panda: A Compiler Framework for Concurrent CPU $$+$$ + GPU Execution of 3D Stencil Computations on GPU-accelerated Supercomputers

Supporting multiple accelerators in high-level programming models

Structure Slicing: Extending Logical Regions with Fields

Contact Info

Product

Resources

About