Achieving Portability and Performance through OpenACC

Herdman, J. A.; Gaudin, Wayne; Perks, Oliver; Beckingsale, David; Mallinson, Andrew C.; Jarvis, Stephen A.

doi:10.1109/waccpd.2014.10

Cited by 22 publications

(11 citation statements)

References 3 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Language extensions focused on exposing parallelism, including OpenMP [15], OpenACC [24], and XScalableMP [33], require user annotations to function. While this allows C programs to efficiently use parallel hardware resources, extensive tuning is necessary to achieve this in practice.…”

Section: Discussion and Related Workmentioning

confidence: 99%

Lifting C semantics for dataflow optimization

Calotoiu

Ben-Nun

Kwasniewski

et al. 2022

Proceedings of the 36th ACM International Conference on Supercomputing

View full text Add to dashboard Cite

C is the lingua franca of programming and almost any device can be programmed using C. However, programming modern heterogeneous architectures such as multi-core CPUs and GPUs requires explicitly expressing parallelism as well as device-specific properties such as memory hierarchies. The resulting code is often hard to understand, debug, and modify for different architectures. We propose to lift C programs to a parametric dataflow representation that lends itself to static data-centric analysis and enables automatic high-performance code generation. We separate writing code from optimizing for different hardware: simple, portable C source code is used to generate efficient specialized versions with a click of a button. Our approach can identify parallelism when no other compiler can, and outperforms a bespoke parallelized version of a scientific proxy application by up to 21%.

show abstract

Section: Discussion and Related Workmentioning

confidence: 99%

Lifting C semantics for dataflow optimization

Calotoiu

Ben-Nun

Kwasniewski

et al. 2022

Proceedings of the 36th ACM International Conference on Supercomputing

View full text Add to dashboard Cite

show abstract

“…Competing approaches including OpenACC [42] and OpenCL are available but have not been as widely adopted so far. For the parallel GPU implementation presented in this paper, we have chosen CUDA because it is the de-facto standard for GPU-accelerated code and is widely supported.…”

Section: Cudamentioning

confidence: 99%

A GPU-Based Kalman Filter for Track Fitting

Mania

Gray

et al. 2021

Comput Softw Big Sci

View full text Add to dashboard Cite

Computing centres, including those used to process High-Energy Physics data and simulations, are increasingly providing significant fractions of their computing resources through hardware architectures other than x86 CPUs, with GPUs being a common alternative. GPUs can provide excellent computational performance at a good price point for tasks that can be suitably parallelized. Charged particle (track) reconstruction is a computationally expensive component of HEP data reconstruction, and thus needs to use available resources in an efficient way. In this paper, an implementation of Kalman filter-based track fitting using CUDA and running on GPUs is presented. This utilizes the ACTS (A Common Tracking Software) toolkit; an open source and experiment-independent toolkit for track reconstruction. The implementation details and parallelization approach are described, along with the specific challenges for such an implementation. Detailed performance benchmarking results are discussed, which show encouraging performance gains over a CPU-based implementation for representative configurations. Finally, a perspective on the challenges and future directions for these studies is outlined. These include more complex and realistic scenarios which can be studied, and anticipated developments to software frameworks and standards which may open up possibilities for greater flexibility and improved performance.

show abstract

“…This will lead to reduce the programmer workload and improve programmer productivity and achieving better performance than OpenCL and CUDA. [10]  OpenACC allows users to specify three levels of parallelism by using three clauses:…”

Section: Openaccmentioning

confidence: 99%

Parallel Hybrid Testing Tool for Applications Developed by Using MPI + OpenACC Dual-Programming Model

Alghamdi¹,

Eassa²

2019

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

Building massively parallel applications has become increasingly important with coming Exascale related technologies. For building these applications, a combination of programming models is needed to increase the system's parallelism. One of these combinations is the dual-programming model (MPI+X) which has many structures that increase parallelism in heterogeneous systems that include CPUs and GPUs. MPI + OpenACC programming model has many advantages and features that increase parallelism with respect heterogeneous architecture and support different platform with more performance, productivity, and programmability. The main problem in building systems with different programming models that it is a hard job for programmers and it is more error-prone, which is not easy to test. Also, testing parallel applications is a difficult task, because of the non-determined behavior of the parallel application. Even after detecting the errors and modifying the source code, it is not easy to determine whether the errors have been corrected or remain hidden. Furthermore, integrating two different programming models inside the same application makes it even more difficult to test. Also, the misusage of OpenACC can lead to several run-time errors that compilers cannot detect, and the programmers will not know about them. To solve this problem, we proposed a parallel hybrid testing tool for detecting run-time errors for systems implemented in C++ and MPI + OpenACC. The hybrid techniques combine static and dynamic testing techniques for detecting real and potential run-time errors by analyzing the source code and during run time. Using parallel hybrid techniques will enhance the testing time and cover a wide range of errors. Also, we propose a new assertion language for helping in detecting potential run-time errors. Finally, to the best of our knowledge, identifying and classifying OpenACC errors has not been done before, and there is no parallel testing tool designed to test applications programmed by using the dualprogramming model MPI + OpenACC or the single-programming models OpenACC.

show abstract

Achieving Portability and Performance through OpenACC

Cited by 22 publications

References 3 publications

Lifting C semantics for dataflow optimization

Lifting C semantics for dataflow optimization

A GPU-Based Kalman Filter for Track Fitting

Parallel Hybrid Testing Tool for Applications Developed by Using MPI + OpenACC Dual-Programming Model

Contact Info

Product

Resources

About