6th International Workshop on High-Level Parallel Programing Model and Suppoertive Enviroments (HIPS'01)
Venue: San Francisco
|Event Date/Time: Apr 23, 2001||End Date/Time: Apr 27, 2001|
Papers due: tentatively October 30, 2000 (12 midnight MET == 6pm EST)
Author notification: December 18, 2000
Camera-ready final papers due: January 15, 2001
HIPS'01 is a full-day workshop to be held at the IPDPS 2001 focusing on high-level programming of networks of workstations, computing clusters and of massively-parallel machines. Its goal is to bring together researchers working in the areas of applications, language design, compilers, system architecture, and programming tools to discuss new developments in programming such systems.
In recent years, several standards have emerged with an increasing demand of support for parallel and distributed processing. On one end, message-passing frameworks, such as PVM, MPI and VIA, provide support for basic communication. On the other hand, distributed object standards, such as CORBA and DCOM, provide support for handling remote objects in a client-server fashion but also ensure certain guarantees for the quality of services.
The key issues for the success of programming parallel and distributed environments are high-level programming concepts and efficiency. In addition, other quality categories have to be taken into account, such as scalability, security, bandwidth guarantees and fault tolerance, just to name a few.
Today's challenge is to provide high-level programming concepts without sacrificing efficiency. This is only possible by carefully designing for those concepts and by providing supportive programming environments that facilitate program development and tuning.
Past results in parallel computing on one side and distributed systems on the other side present opportunities for an increased transfer of knowledge between the areas. In particular, cluster computing presents a promising framework for parallel computing where advances from distributed systems can be utilized. Achievements in the area of automated performance analysis and performance modelling for parallel systems, on the other hand, may contributed to advances in performance analysis of distributed systems.
Future directions also include alternatives to current standardization practices, for example, by replacing client-server protocols with decentralized ones that may be more suitable for distributed systems. On the other hand, successful programming models, such as the shared-memory paradigm, should be investigated for new trends like cluster computing.
Papers should describe the interaction of high-level programming models with compilers, run time systems, and hardware support. Original submissions are invited in all areas relevant to this theme. Appropriate topics include (but are not restricted to):
Concepts and languages for high-level parallel programming
Concurrent object-oriented programming
Distributed objects and components
Structured parallel programming (skeletons, patterns, ...)
Software engineering principles for parallel system
Automatic parallelization and optimization
High-level programming environments
Automated performance analysis and performance modelling
Debugging techniques and development tools
Distributed shared memory
Implementation techniques for high-level programming models
Operating system support for runtime systems and other middleware
Architectural support for high-level programming models
Guarantees for Quality of Service in distributed environments
Security of communication for distributed execution
Fault tolerance in network computing
Papers should describe the interaction of high-level programming models with compilers, run time systems, and hardware support.
Papers should report new research and should not exceed 3000 words (approximately pages 6 typeset on 16-point spacing), including figures and references. All accepted papers will be presented at the workshop and published as joint proceedings for all IPDPS workshops in the Springer LNCS series.
All papers will be judged on originality, significance, correctness, and clarity. The summary should clearly express the contribution of the paper, both in general and in technical terms. It is essential to identify what was accomplished, explain its significance, and include a comparison with previous work. Authors should make every effort to make the technical content of their papers understandable to a broad audience. If any author has published or presented on a related topic in a journal or a previous conference, the summary should explain how it advances such previous work.
Papers must describe work not previously published in refereed venues. Simultaneous submission to HIPS'01 and another publication outlet (conference or journal) will be considered as grounds for rejection.
Submissions consist of a 100-200 word ASCII abstract and a 3000 word paper, not to exceed 6 pages on regular paper or 8 pages in LNCS format, including figures and references. Submissions must be either electronic (encouraged) or postal (discouraged). We strongly encourage authors to use the LNCS guidelines for preparing their papers since this will be a requirement if the paper is accepted.
Electronic submissions must be received by midnight Central European Time, Monday, October 30, 2000 (tentatively).
Submissions should be sent using the automatic upload form
and include your Postscript file, which must be interpretable by Ghostscript. The Postscript must use standard fonts, or include the necessary fonts, and must be prepared for USLetter (8.5"x11") or A4 page sizes (preferably along the LNCS guidelines). Electronic submissions are successful if acknowledged right away by a receipt message.
If the automatic web-based submission does not work for you, then (and only then) use e-mail submission: Send a single e-mail message (up to 1.6MB) to firstname.lastname@example.org (MIME attachments are allowed). The message should contain the information required by the automatic upload form (but in ASCII) and the Postscript summary.
Authors who cannot meet these requirements should submit hardcopy by postal mail instead.
Postal submissions (discouraged) must be sent to the workshop chair Frank Mueller by airmail and must be received on or before October 30, 2000; 20 copies (printed double-sided if possible) must be provided.
These are firm constraints; submissions not meeting the criteria described above will not be considered