A comparative analysis of Kokkos and SYCL as heterogeneous, parallel programming models for C++ applications

Hammond, Jeff R.; Kinsner, Michael; Brodman, James

doi:10.1145/3318170.3318193

Cited by 16 publications

(10 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The recent announcement of an extension to the LLVM/SYCL implementation used in this study to provide support for NVIDIA GPUs is a very much welcomed step in improving the widespread support for SYCL 5 . The success of any programming model lies primarily in the adoption of both users and vendors in order to establish a well supported ecosystem across many problem domains and the increased support is a positive step.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the performance of HPC-style SYCL applications

Deakin

McIntosh–Smith

2020

Proceedings of the International Workshop on OpenCL

View full text Add to dashboard Cite

SYCL is a parallel programming model for developing single-source programs for running on heterogeneous platforms. To this end, it allows for one code to be written which can run on a different architectures. For this study, we develop applications in SYCL which are representative of those often used in High-Performance Computing. Their performance is benchmarked on a variety of CPU and GPU architectures from multiple vendors, and compared to well optimised versions written in OpenCL and other parallel programming models. CCS CONCEPTS• General and reference → Performance; • Computing methodologies → Parallel programming languages; Massively parallel and high-performance simulations; • Software and its engineering → Parallel programming languages.

show abstract

Section: Discussionmentioning

confidence: 99%

“…The study by Hammond et al compared the SYCL and Kokkos programming models with respect to semantics and parallelism, but does not present performance results directly [5].…”

Section: Related Workmentioning

confidence: 99%

Evaluating the performance of HPC-style SYCL applications

Deakin

McIntosh–Smith

2020

Proceedings of the International Workshop on OpenCL

View full text Add to dashboard Cite

show abstract

“…There are comparative studies to compare the efficiency of parallel-computation platforms [18][19][20][21]. According to [18], SYCL takes longer time to execute common midpoint operation, compared to OpenCL and OpenMP.…”

Section: Cuda (Compute Unified Device Architecture) Opencl (Open Commentioning

confidence: 99%

“…Additionally, kernel source code sizes are the shortest, showing better programmability than other parallel processing platforms. In [20], another well-known parallel Using OpenCL as a back-end inherits hardware independency. If a new chipset vendor provides its new OpenCL driver, we can naturally execute our SYCL codes on that chipset.…”

Section: Cuda (Compute Unified Device Architecture) Opencl (Open Commentioning

confidence: 99%

“…Additionally, kernel source code sizes are the shortest, showing better programmability than other parallel processing platforms. In [20], another well-known parallel processing platform kokkos, QPhiX (QCD for Intel Xeon Phi and Xeon processors), and QUDA (QCD on CUDA) is demonstrated and performs better than SYCL, showing greater GFLOPS (Giga Floating point Operations Per Second) on Lattice QCD (quantum chromodynamics) operations. For more complicated operations, deep neural network implementation for SYCL (SYCL-DNN) was compared to Intel math kernel library for deep neural networks in [21].…”

Section: Cuda (Compute Unified Device Architecture) Opencl (Open Commentioning

confidence: 99%

See 1 more Smart Citation

Large-Scale Data Computing Performance Comparisons on SYCL Heterogeneous Parallel Processing Layer Implementations

2020

View full text Add to dashboard Cite

Today, many big data applications require massively parallel tasks to compute complicated mathematical operations. To perform parallel tasks, platforms like CUDA (Compute Unified Device Architecture) and OpenCL (Open Computing Language) are widely used and developed to enhance the throughput of massively parallel tasks. There is also a need for high-level abstractions and platform-independence over those massively parallel computing platforms. Recently, Khronos group announced SYCL (C++ Single-source Heterogeneous Programming for OpenCL), a new cross-platform abstraction layer, to provide an efficient way for single-source heterogeneous computing, with C++-template-level abstractions. However, since there has been no official implementation of SYCL, we currently have several different implementations from various vendors. In this paper, we analyse the characteristics of those SYCL implementations. We also show performance measures of those SYCL implementations, especially for well-known massively parallel tasks. We show that each implementation has its own strength in computing different types of mathematical operations, along with different sizes of data. Our analysis is available for fundamental measurements of the abstract-level cost-effective use of massively parallel computations, especially for big-data applications.

show abstract