📄 http:^^www.cs.wisc.edu^~blz-pdn^status.html
字号:
Date: Tue, 05 Nov 1996 00:26:42 GMTServer: NCSA/1.5Content-type: text/htmlLast-modified: Wed, 31 May 1995 20:57:45 GMTContent-length: 4310<html><head><title>Blizzard & Paradyn WWW Status Report</title></head><body><hr><h1>Blizzard and Paradyn:Infrastructure and Scalable Tools for Multi-Paradigm Parallel Computers</h1><h2>WWW Status Report as of January 1995</h2><hr><h3><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><IMG SRC="http://www.cs.wisc.edu/~blz-pdn/cow3.gif" ALT="Wisconsin COW" ALIGN=MIDDLE>Wisconsin COW</h3><h3>Recent Accomplishments:</h3><ul><li>Acquired and deployed a Cluster of Workstations (COW) consisting of 40dual-processor Sun SPARCstation20s workstations (see photo). The COWand our previously-acquired Thinking Machines CM-5 allow us to developsolutions that scale from off-the-shelf hardware to large parallelcomputers.<li>Implemented a preliminary version of shared memory across COWworkstation nodes that supports coarse-grain sharing as a precursor toour upcoming fine-grain shared memory system.<li>Implemented a preliminary version of "dynamic instrumentation" thatallows the focus of measurement to change as different "informationtraffic jam" locations are investigated.</ul><address>Last Update: January 1994; Contact Barton Miller (<kbd>bart@cs.wisc.edu</kbd>)</address><h3>Background:</h3>The national defense and other key sectors--such as education, healthcare, and commerce--increasingly benefit from and relying upon massivecomputational power. Parallel computers constructed from an array ofconventional computers offer a cost-effective approach to buying CPUcycles. Although the cost advantage is clear, getting the computersto work together effectively requires better methods for:<ol><li>cooperatively sharing information, and<li>identifying and mitigating "information traffic jams."</ol><h3>Goals:</h3>This project addresses both concerns. Effective information sharingmust permit programmers to tailor a program's sharing to the problembeing solved--rather than being constrain by the limit options builtinto a system. Our system supports message passing (like postalmail), fine-grain shared memory (like reading over each other'sshoulders), and hybrid combinations of the two. Our system can beimplemented on low- to high-end parallel computers, so programs can bereused across a wide range of system. For more information on thispart of the project, <!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><A href="http://www.cs.wisc.edu/~wwt">clickhere.</A><p>The second goal of project is to improve techniques for mitigatinginformation traffic jams--what computer specialists call "performancedebugging." Traditional tools for performance debugging have notscaled to large parallel machines because it is difficult to decide inadvance what to measure and in what detail. Our tools transcend thisproblem by supporting "dynamic instrumentation" that allowsmeasurement activity to adapt to the tool's current hypothesis for theinformation traffic jam's location. For more information on this partof the project, <!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><A href="http://www.cs.wisc.edu/~paradyn">clickhere.</A><h3>Participants & Support:</h3>The principal investigators are<!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><A HREF="http://www.cs.wisc.edu/~bart/bart.html">Barton P. Miller,</a><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><A HREF="http://www.cs.wisc.edu/~markhill/markhill.html">Mark D. Hill,</a><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><A HREF="http://www.cs.wisc.edu/~larus/larus.html">James R. Larus,</a> and<!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><A HREF="http://www.cs.wisc.edu/~david/david.html">David A. Wood</a>in <!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><a href="http://www.cs.wisc.edu/"> Computer Sciences </a>at the <!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><A HREF="http://wiscinfo.wisc.edu">University of Wisconsin.</A><p>This work is supported in part by Wright Laboratory Avionics Directorate,Air Force Material Command, USAF, under grant #F33615-94-1-1525 andARPA order no. B550, NSF PYI/NYI Awards CCR-9157366, MIPS- 8957278,and CCR-9357779, NSF Grants CCR-9101035 and MIP- 9225097, DOE GrantDE-FG02-93ER25176, University of Wisconsin Graduate School Grant,Wisconsin Alumni Research Foundation Fellowship and donationsfrom A.T.&T. Bell Laboratories, Digital Equipment Corporation, SunMicrosystems, Thinking Machines Corporation, and Xerox Corporation. OurThinking Machines CM-5 and Cluster of Sun Workstations were purchasedthrough NSF Institutional Infrastructure Grant No. CDA-9024618 withmatching funding from the University of Wisconsin Graduate School.<p><address>Last Update: January 1994; Contact Barton Miller (<kbd>bart@cs.wisc.edu</kbd>)</address></body></html>⌨️ 快捷键说明
复制代码
Ctrl + C
搜索代码
Ctrl + F
全屏模式
F11
切换主题
Ctrl + Shift + D
显示快捷键
?
增大字号
Ctrl + =
减小字号
Ctrl + -