🚀 go-pugleaf

In <37orpd$3jk@ixnews1.ix.netcom.com> DennisR@ix.netcom.com (Dennis Reinhardt) writes: 

>
>                                                                                         9-3-94
>
>Thursday Oct 27
>SCV Computer Society
>
>Subject: Trends In  High Performance Computing
>Speaker: Dr. John Mashey (Silicon Graphics Inc.)
>Time: hors d'oeuvres 7:00 p.m., presentation 7:30 p.m.
>Location: Room 380 C (Math Dept.basement) north corner 
>          Stanford Univ. Quad
>Reservations: not required.
>
>               -------------------
>
>	Trends In  High Performance Computing 
>
>  Dr. John Mashey
>  Silicon Graphics Inc.
>
>This talk will examine fundamental trends in high performance
>computing.  On the hardware side, this technology is characterized
>by clock rates, transistors/chip, SRAM, DRAM, bandwidths, latencies, 
>etc.  On the software side these trends are characterized by
>vectorizing, parallelizing compilers, cache-blocking, etc.  These 
>trends form the basis for comparing different ways to do
>high-performance computing such as multiple interconnected micros or 
>small number of tightly coupled vectorizing uniprocessors and their
>associated memory/cache architectures.  Finally, the talk will look at
>some of the issues in benchmarking high-performance systems.
>
>Dr. Mashey is Director, Systems Technology in Silicon Graphics R&D. His 
>Unix experience dates from his prior employment at Bell Labs starting in 
>1973.  His project experience includes commercial and technical computing 
>both as worker and manager.  He helped design the MIPS RISC 
>architecture.  He has chaired technical conferences on operating systems 
>and CPU chips, most recently Hot Chips VI.  He was a founder of the SPEC 
>benchmarking group, an ACM National Lecturer for 4 years, and Guest 
>Editor for IEEE Micro.  He has given more than 500 public talks on 
>software engineering, RISC design, performance benchmarking, and 
>supercomputing.
>
>Food service starting with this meeting is undergoing a major upgrade:
>catered hors d'oeuvres starting at 7:00.  
>

Two mail messages have asked the same question, so let me clarify: 
this talk is open to the public.  You do not need to be a IEEE CS
member.  .... and it is free of charge.