This is the February 2003 Digest of SIGARCH Messages (sigarch-feb03): * Obituary for Bob Rau http://www.hpl.hp.com/news/2003/jan_mar/rau.html Submitted by Norm Jouppi <Norm.Jouppi@hp.com> * PACT'03 Call for Papers http://www.pactconf.org Submitted by David Kaeli <kaeli@ece.neu.edu> * Hot Chips 15 Call for Contributions http://www.hotchips.org Submitted by Alan Smith <smith@eecs.berkeley.edu> * Latest papers in Computer Architecture Letters http://www.comp-arch-letters.org Submitted by Kevin Skadron <skadron@cs.virginia.edu> --Doug Burger SIGARCH Information Director infodir_SIGARCH@acm.org * Archive: http://www.cs.wisc.edu/~lists/archive/sigarch-members/maillist.html * Web pages: http://www.cs.wisc.edu/~arch/www/, http://www.acm.org/sigarch/ * To remove yourself from the SIGARCH mailing list: mail listserv@acm.org with message body: unsubscribe SIGARCH-MEMBERS ----------------------------------------------------------------- Doug Burger Office: 3.432 ACES Assistant Professor Phone: 512-471-9795 Department of Computer Sciences Assistant: 512-471-9442 University of Texas at Austin Fax: 512-232-1413 Taylor Hall 2.124 E-mail: dburger@cs.utexas.edu Austin, TX 78712-1188 USA www.cs.utexas.edu/users/dburger ----------------------------------------------------------------- ---------------------------------------------------------------------- ---------------------------------------------------------------------- * Obituary for Bob Rau As many of you have heard, Bob Rau, a pioneer in our field, passed away recently. His passing is a tremendous loss for our community. His obituary can be found at the following location: http://www.hpl.hp.com/news/2003/jan_mar/rau.html ---------------------------------------------------------------------- ---------------------------------------------------------------------- * PACT'03 Call for Papers PACT'03 12th International Conference on Parallel Architectures and Compilation Techniques September 27 - October 1, 2003 Chateau Sonesta Hotel, New Orleans, LA http://www.pactconf.org ***************************************************************** PACT-03 will be held in New Orleans, LA, a major cultural center of the southern United States. The city offers such attractions as Mardi Gras, the French Quarter, Bourbon St., the Mississippi River and is the home of dixieland jazz. PACT is a multi-disciplinary conference that brings together researchers from the hardware and software areas to present ground-breaking research related to parallel systems ranging across instruction-level parallelism, thread-level parallelism, multiprocessor parallelism and distributed systems. PACT solicits papers on advances in architecture, compilers, languages and applications across a broad range of topics, including, but not limited to: - Hardware/software optimizations for memory hierarchies - Programming languages for parallel scientific and object-oriented applications - Superscalar, VLIW, and multithreading architectures - I/O, network processing and O/S issues for parallel computing - Parallel aspects of power-aware and mobile/wireless computing - Reconfigurable computing and novel parallel architectures - Just-in-time/dynamic compilation for parallelism - Parallel algorithms, computation models and simulation techniques - Parallel software development tools supporting performance analysis/tuning and debugging Selected papers will be invited for publication in the Journal of Instruction Level Parallelism. Please check the following web site for paper submission information: http://www.pactconf.org *********************************************** Important Deadlines Abstract Submission Deadline: April 4, 2003 Full Paper Submission Deadline: April 11, 2003 Author Notification: June 20, 2003 Final Papers Due: July 18, 2003 *********************************************** General Chairs David Kaeli, Northeastern University David Koppelman, LSU Program Chairs Mary Hall, USC/ISI Vivek Sarkar, IBM Local Arrangements J. Ramanujam, LSU Finance Chair Nikos Pitsianis, Duke University Registration Chair Diana Keen, Cal Poly Tutorial Chair Csaba Andras Moritz, U. of Massachusetts Workshop Chair Martin Schulz, Cornell University Publications Chair Bruce Childers, University of Pittsburgh Publicity Chair Dieter Kranzlmueller, Kepler University Website Chair Gene Cooperman, Northeastern University Josep Llosa, UPC Barcelona Travel Awards Chair Lizy John, University of Texas, Austin ---------------------------------------------------------------------- ---------------------------------------------------------------------- * Hot Chips 15 Call for Contributions HOT Chips 15 CALL FOR CONTRIBUTIONS Stanford University Palo Alto, California August 17-19, 2003 Since it began in 1989, Hot Chips has been known as one of the semiconductor industry's leading conferences on high-performance microprocessors and related integrated circuits. The conference is held once a year in August on the Stanford University campus in the center of the world's capital of electronics activity, Silicon Valley. The emphasis this year, as in previous years, is on real products and realizable technology. Topics of interest for this year's conference include but are not limited to: * Microprocessors * Systems-on-chip * Embedded processors * Low-power chips * Digital signal processors * Performance evaluation * Dynamic power management * Application-specific processors * Network/security processors * Novel chips: quantum computing, microarray * Graphics/Multimedia/Game processors * Communication/networking chips * Wireless LAN/Wireless WAN chips * Reliability and design for test * Compiler technology * Operating system/chip interaction * Reconfigurable processors * Advanced semiconductor process technology Presentations at HOT Chips are in the form of 30-minute talks. Presentation slides will be published in the HOT Chips Proceedings. Participants are not required to submit written papers, but a select group will be invited to submit a paper for inclusion in a special issue of IEEE Micro. Submissions must consist of a title, extended abstract (three pages maximum), and the presenter's contact information (name, affiliation, job title, address, phone, fax, and email). Please indicate whether you have submitted, intend to submit or have already presented or published a similar or overlapping submission to another conference or journal. Also indicate if you would like the submission to be held confidential; we do our best to maintain confidentiality. Submissions are evaluated by the Program Committee on the basis of the performance of the device (for devices), degree of innovation, use of advanced technology, potential market significance and anticipated interest to the audience. Research and software contributions will be evaluated with similar criteria. Authors will be notified of the status of their submission by the end of April, 2003. Don't miss this chance to present your work to an audience of engineers, computer architects, and computer system and device researchers. Submissions must be received no later than March 15, 2003. Please make your submissions in plain ascii text (in the message, not as an attachment) to: hotchips-submission@arith.stanford.edu (Submissions containing figures may be submitted in pdf, but plain ascii is preferred.) For more information, see the Hot Chips 15 Web site at: http://www.hotchips.org Send questions to hotchips@arith.stanford.edu or contact the co-program chairs: Prof. Michael Flynn at flynn@ee.stanford.edu, or Pradeep K. Dubey at pdubey@broadcom.com, 408-922-5904. Sponsored by the Technical Committee on Microprocessors and Microcomputers of the IEEE Computer Society See the HOT CHIPS 15 web page for updates: http://www.hotchips.org ---------------------------------------------------------------------- ---------------------------------------------------------------------- * Latest papers in Computer Architecture Letters Computer Architecture Letters is pleased to announce the publication of two more papers online at our website, <http://www.comp-arch-letters.org>; abstracts appear below. The papers will appear in print in our next paper issue. The print issues are distributed to the entire IEEE Computer Society TCCA membership, and e-mail notifications of newly accepted papers are sent on a regular basis to the TCCA and ACM SIGARCH memberships. Papers: - YC. Sohn, NH. Jung, SR. Maeng. "Request Reordering to Enhance the Performance of Strict Consistency Models." - K. A. Shaw, W. J. Dally. "Migration in Single Chip Multiprocessors." The objective of Letters is to publish short (4-page), timely articles of high-quality work. We are very much aware of the long delays in our field between submissions of manuscripts and their eventual appearance in print. We are doing something about that with this journal. After just over one year of operation, we have maintained an average turnaround time from submission to author notification of just one month, with an acceptance rate of 21%. We encourage the community to continue submitting papers to Letters. Submissions are welcomed on any topic in computer architecture, especially but not limited to: - Microprocessor and multiprocessor systems - Microarchitecture and ILP processors - Workload characterization - Performance evaluation and simulation techniques - Compiler-hardware and operating system-hardware interactions - Interconnect architectures - Memory and cache systems - Power and thermal issues at the architecture level - I/O architectures and techniques - Independent validation of previously published results - Analysis of unsuccessful techniques - Network and embedded-systems processors - Real-time and high-availability architectures - Reconfigurable systems The call for papers and instructions for submission can be found at <http://www.comp-arch-letters.org> Abstracts --------- YC. Sohn, NH. Jung, SR. Maeng. "Request Reordering to Enhance the Performance of Strict Consistency Models." Abstract: Advances in ILP techniques enable strict consistency models to relax memory order through speculative execution of memory operations. However, ordering constraints still hinder the performance because speculatively executed operations cannot be committed out of program order for the possibility of mis-speculation. In this paper, we propose a new technique which allows memory operations to be non-speculatively committed out of order without violating consistency constraints. K. A. Shaw, W. J. Dally. "Migration in Single Chip Multiprocessors." Abstract: Global communication costs in future single-chip multiprocessors will increase linearly with distance. In this paper, we revisit the issues of locality and load balance in order to take advantage of these new costs. We present a technique which simultaneously migrates data and threads based on vectors specifying locality and resource usage. This technique improves performance on applications with distinguishable locality and imbalanced resource usage. 64% of the ideal reduction in execution time was achieved on an application with these traits while no improvement was obtained on a balanced application with little locality. ---------------------------------------------------------------------- ----------------------------------------------------------------------