Text

Scope and Interests

The technological trends of HPC system evolution indicates increasing burden for application developers in management of the unprecedented levels of complexity in hardware and the associated performance characteristics. Many existing application codes are unlikely to perform well on future systems without major modifications or even complete rewrites. It will be important to utilize, in unison, many characteristics such as multiple levels of parallelism, many lightweight cores, complex memory hierarchies, novel I/O technology, power capping, systemwide temporal/spatial performance heterogeneity and reliability concerns. The HPC community has developed new programming models, algorithms, libraries and tools to meet these challenges in order to accommodate productive code development and effective system use. However, the application community still needs to identify the benefit through practical evaluations.

Thus, the focus of this workshop is on methodologies and experiences of scientific and engineering applications and algorithms to achieve sustainable code development for better productivity, application performance and reliability. In particular, we will focus on the following topics in parallel and distributed scientific and engineering applications, but not limited to:

• Code modernization methodologies and experiences for adapting the changes in future computing systems such as porting of legacy simulation code and libraries/tools to facilitate code refactoring and porting.
• Application and algorithm development of various parallel and distributed programming models/framework such as CAF, UPC, Chapel, X10, Charm++, HPX, Uintah, Legion, and/or the interoperation of multiple models within single applications (e.g. MPI+X where X is OpenMP, OpenCL, CUDA etc). We appreciate the experiences of early adopters of new programming models and platforms.
• Experience in new tools and libraries for effective application development, including performance tools, application development frameworks, Domain Specific Languages (DSLs), etc.
• Tools and techniques for improving application reliability and resilience. This includes both performance and correctness issues, with the latter arising from adverse operating conditions (e.g. low power) or very large system scales.
• Use cases of enterprise distributed computing technology (such as MapReduce, Data Analytics and Machine-learning tools) in scientific and engineering applications.
• Large-scale parallel and distributed algorithms supporting science and engineering applications.
• Methodologies and experiences in developing large-scale applications.

Publications

The conference proceedings will be published by IEEE Computer Society. Selected papers from the proceedings will be invited for publication in journal special issues.

Important Dates

Paper Submission Guidelines

All papers must be submitted electronically and in PDF format. The material presented should be original and not published or under submission elsewhere. Authors should submit full papers of up to 10 pages, following strictly the IEEE Computer Society Proceedings Manuscript style, using two-column, single-space format, with 10-point font size. Figures and references must be included in the 10 pages. Oversized papers will be automatically rejected by the PC chairs. At least one of the authors of each accepted paper must register early to attend the conference, in order for the paper to appear in the conference proceedings.

PDSEC-2018 paper submission link: https://easychair.org/conferences/?conf=pdsec2018