|
The 2000 International Symposium on
|
ACM SIGPLAN Co-located with OOPSLA 2000 |
|
Call for Participation and Advance ProgrammeThe International Symposium on Memory Management is a forum for research in memory management, in all its diversity. Areas of interest include but are not limited to: garbage collection, dynamic storage allocation, storage management implementation techniques and their interactions with languages and operating systems, and empirical studies of programs' memory allocation and referencing behavior. ISMM 2000 continues the tradition of the successful conference series established with the International Workshops on Memory Management held in 1992 (St. Malo, France) and 1995 (Kinross, Scotland), and the inaugural ISMM in 1998 (Vancouver, Canada). Proceedings of the IWMM conferences are available from Springer-Verlag (Lecture Notes in Computer Science no. 637 and no. 986), while the ISMM'98 proceedings were published by the ACM. The 2000 Symposium Proceedings are also published by the ACM.
|
Sunday October 15th | Monday October 16th | ||
8:15-9:00 | Coffee & Registration* | 8:15-9:00 | Coffee |
9:00-9:15 | Welcome & Introduction | 9:00-10:20 | Session III: HardwareSupport |
9:15-10:20 | Invited Speaker: Jon L. White, CommerceOne, Inc. | ||
10:20-10:50 |
Break
|
10:20-10:50 |
Break
|
10:50-12:30 | Session I: Accuracy & Locality | 10:50-12:30 | Session IV: Profiling & Object Lifetimes |
12:30-14:00 |
Lunch
|
12:30-14:00 |
Lunch
|
14:00-15:00 | Proposal & Discussion: Trace Formats | 14:00-15:00 | Informal Presentations |
15:00-15:30 |
Break
|
15:00-15:30 |
Break
|
15:30-17:00 | Session II: Implementation | 15:30-17:00 | Session V: Concurrent & Distributed |
17:30-18:30 |
OOPSLA tutorial/workshop reception
|
17:30-19:30 |
OOPSLA welcome reception
|
18:30-20:30 |
Conference dinner
|
*Registration will also be available from 7am to 5pm on Sunday, and 7am to 2pm on Monday.
Conquering the Hurdles, Challenging the Horizons A Historical Perspective on Memory Management
We will travel through a retrospective overview of some of the great "hurdles" in the area of Memory Management, from the discovery of "Garbage Collection", through the realization of the potential harmfulness of GC, to the expanded horizons which are driven by paradigm shifts in application programs. In the process, we will review some of the very early approaches to memory management to appreciate just how far we have come, and just how much the great leaps of this science have been driven by the appearance of cheaper and faster hardware. We will also re-visit some "war stories" of the past where a paradigm shift was forced because of unforseen consequences in what was otherwise perceived to be a "Very Good Idea(tm)". Examples will include the speaker's own pioneering experiences in shifting from pure "collection" techniques to BIBOP storage strategies, compile-time change-of-representation for a one to two orders of magnitude speedup in numerical programs written in Lisp, to the discovery of the "Pig-in-a-Python" syndrome. Speculative, overview-like conjectures will be made about the current horizons of the field, especially with respect to large databases and the Java programming language.
A Brief Biography of Jon L. White
JonL received a B.S. degree in Mathematics from Carnegie-Mellon University, and an M.A. in Applied Mathematics from Harvard University. He subsequently joined Marvin Minsky's team at the M.I.T Artifical Intelligence Laboratory, where he began the development of high-quality Lisp systems and compilers. The PDP10 MacLisp efforts subsequently supported not only the A.I. lab, but the MACSYMA Symbolic Algegra group, and ultimately many other university and commercial research labs. JonL also spent one calendar year away from MIT at IBM's Watson Research Center in Yorktown Heights, NY, contributing to the development of Lisp/370.
At the birth of the commercial A.I. revolution, JonL joined a group at Xerox's Palo Alto Research Center tasked with product development of Interlisp on the special-purpose, microded Xerox D-series machines. A few years later, in January 1985, he joined the "startup" company Lucid, Inc. whose goal was to make high-quality Common Lisp systems available on all the prevalent "stock" hardware machines. As the commercial A.I. market - and consequently the commercial Common Lisp market - went into decline in the early 1990's, JonL joined the Lisp group at Harlequin. In all of these commercial Lisp developments, JonL contributed substantially not only to the compiler development, but also to both "GC" and higher-level memory management.
After a brief 8 months at NASA/Ames Research Center working with the Automated Software Engineering group there, JonL joined yet another startup - a spin-off from Stanford University - which was quickly bought out by his present employer CommerceOne. He is currently working on a product called iMerge, written in Lisp and a result of many years of A.I. research at Stanford which is directed at the database needs of today's Business-to-Business commercial environment.
General Chair | Programme Chair |
Craig Chambers
Department of Computer Science and Engineering University of Washington Box 352350 Seattle, WA 98195-2350 USA +1 (206) 685-2094; fax: +1 (206) 543-2969 chambers@cs.washington.edu |
Tony Hosking
Department of Computer Sciences Purdue University 1398 Computer Science Building West Lafayette, IN 47907-1398 USA +1 (765) 494-6001; fax: +1 (765) 494-0739 hosking@cs.purdue.edu |
Steering Committee | Programme Committee |
Hans Boehm
Hewlett-Packard |
Trishul Chilimbi
Microsoft Research, USA |
Richard Jones University of Kent at Canterbury, UK |
Bart Demoen Katholieke Universiteit Leuven, Belgium |
J. Eliot B. Moss University of Massachusetts at Amherst, USA |
David Detlefs Sun Microsystems Laboratories, USA |
Simon Peyton Jones Microsoft Research, UK |
Eric Jul
University of Copenhagen, Denmark |
Ben Zorn Microsoft Research, USA |
Greg Morrisett
Cornell University, USA |
Scott Nettles
University of Texas at Austin, USA |
|
Marc Shapiro
Microsoft Research, UK |
Maintained by Richard Jones