There are two significant challenges when programming for today’s GPU-accelerated supercomputers: 1). Different platforms consist of accelerators from different vendors, 2). Compute nodes often feature multiple accelerators. Given the trend of very fat compute nodes, it is more important than ever to make effective use of the increased acceleration potential. It is also highly desirable that application programmers can use a single programming approach to target the architectural components of a single compute node and across different supercomputing platforms. Therefore, the focus of this workshop is on the use of high-level programming methods that can exploit the maximum compute capability out of such systems. Historically, the favored portable approaches were OpenMP offloading and OpenACC, both based on directives. Today, we recognize the evolution of other options to adapt to heterogeneity and, starting in 2021, we extended the workshop scope to include use of Fortran/C++ standard language parallelism, SYCL, DPC++, Kokkos, RAJA as well as task-based and data-centric models like Regent, Legion, OmpSs among several alternatives that can provide scalable as well as portable solutions without compromising on performance. A programmer’s expectation from the software community is to deliver solutions that would allow maintenance of a single code base whenever possible, thus avoiding duplicate effort across programming models and architectures.
The 10th WACCPD workshop will highlight the improvements over the state of art through accepted papers and prompt discussion through keynotes and panels. The workshop aims to showcase all aspects of accelerator programming for heterogeneous systems such as innovative high-level language or library approaches, lessons learned while using directives or other portable approaches to migrate scientific legacy code to modern systems, and compilation and runtime scheduling techniques.
In the past nine years of this workshop, WACCPD has been one of the major forums at the Supercomputing Conference to bring together programming model users, developers, and the tools community to share knowledge and experiences to tackle emerging complex parallel computing systems.
Topics of interest for workshop submissions include (but are not limited to)
- Application developer’s experiences porting scientific applications to modern systems
- Critical assessment of the possibilities and limitations of minimal-maintenance approaches with respect to achieved portability and performance — Support the analysis with case studies
- Compiler and runtime support for current and emerging architectures (e.g. heterogeneous architectures, low-power processors)
- Evaluation of alternative abstractions for expressing parallelism or managing complex/heterogeneous memory hierarchies in base languages, frameworks or directives-based programming languages
- Extensions to and shortcomings of current directives for heterogeneous systems
- Auto-tuning and optimization strategies leveraging high-level programming abstractions
- Extending directive-based approaches to other environments such as Julia or Python
- Parallel computing using hybrid programming paradigms (e.g. MPI, OpenMP, OpenMP offloading, OpenACC, OpenSHMEM, OneAPI, OmpSs, SYCL, DPC++, Kokkos, RAJA, Regent, Legion, C++/Fortran standard language parallelism)
- Asynchronous execution and scheduling (task-based approaches)
- Power/energy studies and solutions targeting accelerators or heterogeneous systems
Workshop Important Deadlines
- Paper Submission Deadline:
August 4, 2023August 11, 2023August 17, 2023 (final extension) 11:59pm (UTC-12) - Author Notification: September 8, 2023
- Camera Ready Deadline: September 29, 2023
Best Paper Award
The Best Paper Award will be selected on the basis of explicit recommendations of the reviewers and their scoring towards the paper’s originality and quality. In order to be considered for the Best Paper Award, the authors must submit an Artifact Description appendix according to the reproducibility initiative of the SC23 technical papers.
Paper Proceedings
WACCPD papers will be peer-reviewed and selected for presentation at the workshop. The papers presented will be published in the SC23 Workshop Proceedings.
Authors must use the new proceeding templates and consult the CCS2012 guide available at ACM Proceedings Template. Authors may use the proceedings templates, either for LaTeX or for Word, for the preparation of their papers.
Paper Submission Guidelines
- For the first time WACCPD paper submissions and workshops talks are organized in two separate tracks:
- Research Papers Track
- Submissions between 6 to 10 pages including references and figures
- 30-minute talk (25 minute presentation + 5 minute Q&A)
- Early Career Track
- Available to academic, industry, and laboratory staff and post-docs within the first five years of a permanent position
- Submissions between 2 and 4 pages including references and figures
- 10 minute lightning talk (5 minute presentation + 5 minute Q&A)
- Research Papers Track
- Papers should be submitted electronically via the SC23 Submission Page.
- They must follow the proceeding templates as described in https://www.acm.org/publications/proceedings-template.
- Page limits does not include appendix pages for describing data artifacts and reproducibility actions as outlined in the reproducibility initiative of this workshop.
- Submitted papers should not have appeared in or be under consideration for a different workshop, conference or journal.
- In submitting the paper, the authors acknowledge that at least one author of an accepted submission will register for and attend the workshop.
- Camera ready submissions can have up to 12 pages in Research Papers Track and 4 pages in Early Career Track (including references and figures) to accommodate feedback from reviewers.
- Reproducibility files will be uploaded to Linklings with the submission.
Reproducibility
WACCPD supports SC23’s reproducibility initiative. In contrast to the SC23 technical paper submission process, WACCPD is not requiring reproducibility information, however, strongly encourages submitters to provide artifact appendices in form of an Artifact Description (AD) and eventually artifacts that can complete an Artifact Evaluation (AE).
Authors should follow the instructions on SC23’s reproducibility initiative. Similar to SC23’s reproducibility process, AD/AE appendices will be automatically-generated from an AD/AE appendix form provided on the WACCPD submission page. Authors are encouraged to familiarize themselves early with the AD/AE appendix requirements on the SC23 web page and the WACCPD’s submission form.