Publication | Closed Access
Proceedings of the 20th annual international conference on Supercomputing
25
Citations
0
References
2006
Year
Unknown Venue
Massively-parallel ComputingExascale ComputingOpen Source SupercomputingEngineeringComputer EngineeringComputer ArchitectureComputing SystemsParallel ProgrammingComputer ScienceAcm International ConferenceYear Google SearchParallel ComputingSupercomputer ArchitectureNext Generation Computing
Welcome to the 20th ACM International Conference on Supercomputing.Much has changed in the last 20 years of supercomputing and if anything the pace quickens. We now have silicon real estate to burn; literally in some cases energy management being a key major research domain within computer and telecommunication engineering disciplines. We also have the ability to explore, and in some cases re-explore, architectural options which seemed far-fetched when ICS commenced; far-fetched even for those who have been around that long and who still suffer from incurable optimism. If we look back a little further than a 2 year Google search there are many interesting things to rediscover.We do hope you will take the opportunity to explore a little wider than Cairns in your visit to Australia. For some of you this will be your second ICS with the last being in Melbourne in what was mid winter! Australia is a very big country and the Queensland weather and scenery are a little different to that of Melbourne.Welcome to ICS 2006, the ACM International Conference on Supercomputing, in Cairns, Queensland, home of the Great Barrier Reef and Australia's tropical rainforest. Since its inception in 1987, the ICS has been bringing together a rich diversity of authors from universities, government and industry. We hope that this conference, now in its 20th anniversary year, will continue to be a leading forum on all aspects of high performance computing.This year, 141 papers from 21 different countries were submitted to ICS, covering virtually all areas of supercomputing. The paper selection process was handled by an outstanding Technical Program Committee. Each paper was rigorously reviewed by at least three Program Committee members, while some papers received as many as five written reviews. Based on the referees' comments and ratings, the Committee selected 37 papers, which translated to an acceptance rate of 26%. This rate is similar to last year's 27% acceptance rate, ensuring that this conference continues to maintain a very high level of competitive standards.The Technical Program Committee had a strong international flavor, with members coming from 17 different countries in four continents. We had an excellent TPC meeting with high attendance held in Delft, The Netherlands. Some members attended in person, and we arranged for others to attend via phone and net conference. This way we made certain that everyone could participate irrespective of their location.To help the Program Committee members focus their reviewing effort on papers requiring the most attention, the selection procedure included a rebuttal process. This served to provide the authors an opportunity to respond to comments made in the reviews before the program committee meeting was held. The rebuttal process had greatly aided previous conferences. It ensures that the reviewers' claims have been carefully scrutinized, and it helps the organizers to identify borderline papers and collect additional reviews. Due to the large number of papers submitted and the low acceptance ratio, we regret that many high-quality papers could not be accepted. Our hope is that the referees' comments were helpful and that this will only lead to a brief delay before those papers are published somewhere else.As in previous years, the goal of this year's conference is to bring together papers covering widespread topics encompassing architecture, applications, and other related areas in high performance computing. In addition to the high-quality regular paper sessions, there will be two workshops, three instructive tutorials, and two inspiring keynote speeches on both timely and challenging issues in these areas. The topics are from traditionally important supercomputing disciplines as well as newly emerging and exciting areas of research.