[Please accept our apologies if you receive multiple copies of this CFP] ------------------------------------------------------- 1st CALL FOR PAPERS Special Issue on SEMANTICS AND COST MODELS FOR HIGH-LEVEL PARALLEL PROGRAMMING to appear in Computer Languages, Systems and Structures http://f.loulergue.free.fr/CLSS ------------------------------------------------------- AIMS AND SCOPE Many applications of computing require performance levels attainable only on parallel architectures. Such systems are now readily available as their price/performance ratio continues to improve. The rapid development of affordable hardware for parallel computing makes the need to develop high-quality parallel software increasingly urgent. Sequential programming has long benefited from high-level programming techniques and tools that have made today's immense range of software economically viable. Two decades of research into high-level parallel programming has produced methods and tools that improve the price/performance ratio of parallel software, and broaden the range of target applications. GRIDs offer tremendous computing power. Nevertheless, this power is far from being effectively exploited. In addition to technical problems related to portability and access, Grid computing needs new programming paradigms. Research on high-level grid programming is particularly relevant. Expressivity and ease of use are not the only concerns of high-level parallel languages. Languages based on formal semantics allow one to prove the correctness of programs, to refine programs, etc. One can also have more confidence in the safety of the programming and execution environments by proving the correctness of compilation, the correctness of lower-level semantics (e.g. abstract machines) with respect to the high-level semantics, etc. These concerns are shared by sequential languages. In parallel programming the need for performance prediction is especially great. For parallel programming, one must use an efficient parallel algorithm right from the start; very often this has little to do with the corresponding sequential algorithm. Thus cost models are needed to allow discrimination between efficient and inefficient parallel algorithms. Such models must be sufficiently simple, abstract, and closely related to the source code to allow the programmer to understand which aspects of the algorithm must be modified to improve efficiency. TOPICS We welcome submission of original, unpublished papers in English on topics including: o Parallel language design and implementation based on semantics o Semantics and/or cost models of algorithmic skeletons and parallel libraries o Semantics and/or cost models of programming languages for GRID computing o Proofs of correctness of high-level parallel programs o Refinement of high-level parallel programs o Parallel program design and optimization based on cost models o Experimental results which validate cost models PAPER SUBMISSION Prospective authors are invited to submit full papers in English presenting original research. Submitted papers must be unpublished and not submitted for publication elsewhere. Papers will go through a rigorous reviewing process. Each paper will be reviewed by the guest editor and researchers from the high-level parallel programming community. Papers should be sent to loulergue@univ-paris12.fr with subject "[CLSS Special Issue] Submission" no later than Monday 3 November 2003. An early email with your intention to submit a paper would be greatly appreciated (especially if you have doubts about the relevance of your paper). FORMATTING INSTRUCTIONS Formatting instructions for paper submissions will be available at: http://f.loulergue.free.fr/CLSS IMPORTANT DATES November 3, 2003: Full paper due January 15, 2003: Notification April 15, 2004: Final version due Automn 2004: Expected date of publication GUEST EDITOR Dr. Frederic LOULERGUE Laboratory of Algorithms, Complexity and Logic (LACL) University of Paris Val de Marne 61, avenue du General de Gaulle F-94010 CRETEIL - FRANCE loulergue@univ-paris12.fr