http:^^www.cs.washington.edu^homes^grant^work.html
来自「This data set contains WWW-pages collect」· HTML 代码 · 共 176 行
HTML
176 行
Date: Tue, 10 Dec 1996 22:50:38 GMTServer: NCSA/1.4.2Content-type: text/htmlLast-modified: Fri, 18 Oct 1996 18:44:03 GMTContent-length: 7679<html><head> <TITLE>Brian Kris Grant: Work Background</TITLE></head><body bgcolor="#FFFFFF" TEXT="#004535" LINK="#EE0000" VLINK="#952045" ALINK="#000000"><font color="#FF4444"><h1>Work Background</h1></font><h3>Phased Compilation</h3><p><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><!WA0><a href="http://www.cs.washington.edu"><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><!WA1><img border=0 align=rightsrc="http://www.cs.washington.edu/general/sptm_s.gif"></a> I'm inmy third year of graduate school at the <!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><!WA2><ahref="http://www.washington.edu">UW</a>. I am now starting to explorethe new area of <!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><!WA3><a href="http://www.cs.washington.edu/homes/grant/phasedcomp/index.html">phasedcompilation</a>, with <!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><!WA4><ahref="http://www.cs.washington.edu/people/faculty/eggers.html">SusanEggers</a> as my advisor, and am part of the <!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><!WA5><ahref="http://www.cs.washington.edu/research/projects/unisw/DynComp/www/">UWDynamic Compilation Group</a>. Phased compilation encompasses suchideas as <!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><!WA6><ahref="http://www.lpac.ac.uk/SEL-HPC/Articles/GeneratedHtml/functional.partial.html">partialevaluation</a>, incremental specialization, <!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><!WA7><AHREF="http://www.cs.washington.edu:80/homes/pardo/rtcg.d/index.html">run-timecode generation</a>, and <!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><!WA8><ahref="http://www.cs.washington.edu/research/projects/unisw/DynComp/www/">dynamiccompilation</a> (these concepts overlap). I plan to do my generalsand my thesis in this area, and hope to graduate around the time our<!WA9><!WA9><!WA9><!WA9><!WA9><!WA9><!WA9><!WA9><!WA9><!WA9><a href="http://www.cs.washington.edu/building/">new building</a> iscompleted (or soon after).<h3>Other Interests</h3><p>In addition to phased compilation, lately I've been interested in<!WA10><!WA10><!WA10><!WA10><!WA10><!WA10><!WA10><!WA10><!WA10><!WA10><ahref="http://www.cs.washington.edu/research/projects/spin/www/">extensibleoperating systems</a> and <!WA11><!WA11><!WA11><!WA11><!WA11><!WA11><!WA11><!WA11><!WA11><!WA11><ahref="http://www-dsg.stanford.edu/DSGHomePage.html">distributedsystems</a>. This past spring, I read up on techniques for creatingand exploiting <!WA12><!WA12><!WA12><!WA12><!WA12><!WA12><!WA12><!WA12><!WA12><!WA12><a href="http://www.micro29.org/">instruction-levelparallelism (ILP)</a>. Many of the papers came from the <!WA13><!WA13><!WA13><!WA13><!WA13><!WA13><!WA13><!WA13><!WA13><!WA13><AHREF="http://www.crhc.uiuc.edu/Impact/">IMPACT</a> group at <!WA14><!WA14><!WA14><!WA14><!WA14><!WA14><!WA14><!WA14><!WA14><!WA14><ahref="http://www.uiuc.edu">UIUC</a>.<p>In the past I've been interested in <!WA15><!WA15><!WA15><!WA15><!WA15><!WA15><!WA15><!WA15><!WA15><!WA15><AHREF="http://www.umiacs.umd.edu/~dbader/sites.html">parallel</a>programming languages and environments, parallel algorithms, and <!WA16><!WA16><!WA16><!WA16><!WA16><!WA16><!WA16><!WA16><!WA16><!WA16><ahref="http://csep1.phy.ornl.gov/csep.html">scientific computing</a>.I'm also concerned about the <!WA17><!WA17><!WA17><!WA17><!WA17><!WA17><!WA17><!WA17><!WA17><!WA17><a href="http://www.cs.washington.edu/homes/grant/society.html">impact ofcomputing technologies on society</a>.<h3>Collective Communication</h3><p>In spring 1995, I finished my quals project under <!WA18><!WA18><!WA18><!WA18><!WA18><!WA18><!WA18><!WA18><!WA18><!WA18><ahref="http://www.cs.washington.edu/people/faculty/ladner.html">RichardLadner</a>, <em>Design and Analysis of Collective-CommunicationPrimitives</em>. I have a few more ideas that I'm exploring in thisarea and I should have a paper available later this year.<h3>GAPP</h3><p>In my second year, I also did a <!WA19><!WA19><!WA19><!WA19><!WA19><!WA19><!WA19><!WA19><!WA19><!WA19><a href="http://www.cs.washington.edu/homes/grant/552proj.html">coolproject</a> with <!WA20><!WA20><!WA20><!WA20><!WA20><!WA20><!WA20><!WA20><!WA20><!WA20><ahref="http://www.cs.washington.edu/homes/eric">Eric Anderson</a> forthe <!WA21><!WA21><!WA21><!WA21><!WA21><!WA21><!WA21><!WA21><!WA21><!WA21><ahref="http://www.cs.washington.edu/education/courses/552/">distributedand parallel systems course</a>. It was a general-purpose,user-programmable, kernel-resident packet processor (boy, what amouthful!). It was a good idea, but its performance suffered a littledue to a massive, sparse switch statement that was converted to aseries of tests and branches by the compiler. We never had time torewrite it and collect some real times, but we're sure they would havebeen great. Actually, dynamic compilation would have been ideal forthis application. We also had some other ideas for optimizing thecomputation of which packet processors should execute on a givenpacket, but development on <!WA22><!WA22><!WA22><!WA22><!WA22><!WA22><!WA22><!WA22><!WA22><!WA22><ahref="http://www.cs.washington.edu/research/projects/spin/www/papers/Usenix96/extprotarch.ps">SPIN'snetworking framework</a> was beginning by the end of the quarter, sowe didn't bother to pursue it further.<h3>Teaching</h3><p>During my first year at UW, when my <!WA23><!WA23><!WA23><!WA23><!WA23><!WA23><!WA23><!WA23><!WA23><!WA23><ahref="http://www.cs.washington.edu/homes/map/office/225.html">office</a>was in Sieg, I was a teaching assistant, first for <!WA24><!WA24><!WA24><!WA24><!WA24><!WA24><!WA24><!WA24><!WA24><!WA24><ahref="http://www.cs.washington.edu/education/courses/142/">CSE/ENGR-142</a>,the first-semester programming course (when it was in Ada), then for<!WA25><!WA25><!WA25><!WA25><!WA25><!WA25><!WA25><!WA25><!WA25><!WA25><ahref="http://www.cs.washington.edu/education/courses/451/CurrentQuarter/">CSE-451</a>,the undergraduate operating systems course. Now, my office is in <!WA26><!WA26><!WA26><!WA26><!WA26><!WA26><!WA26><!WA26><!WA26><!WA26><ahref="http://www.cs.washington.edu/homes/map/chateau.html">TheChateau</a>, and I am supported by a <!WA27><!WA27><!WA27><!WA27><!WA27><!WA27><!WA27><!WA27><!WA27><!WA27><AHREF="http://stis.nsf.gov/ehr/gerd/gradfell.htm">NSF GraduateFellowship</A>.<h3>Lawrence Livermore National Laboratory</h3><p><!WA28><!WA28><!WA28><!WA28><!WA28><!WA28><!WA28><!WA28><!WA28><!WA28><a href="http://www.llnl.gov"><!WA29><!WA29><!WA29><!WA29><!WA29><!WA29><!WA29><!WA29><!WA29><!WA29><img border=0 align=left src="http://www.cs.washington.edu/homes/grant/llnl.gif"></a>I spent three summers at <!WA30><!WA30><!WA30><!WA30><!WA30><!WA30><!WA30><!WA30><!WA30><!WA30><a href="http://www.llnl.gov">LawrenceLivermore National Laboratory</a>. My first summer at LLNL, I studiedcluster message-passing systems, mainly <!WA31><!WA31><!WA31><!WA31><!WA31><!WA31><!WA31><!WA31><!WA31><!WA31><ahref="http://www.epm.ornl.gov/pvm/">PVM</a>. During the next twosummers, while working on my main project, a parallel climate model, Ilooked into <!WA32><!WA32><!WA32><!WA32><!WA32><!WA32><!WA32><!WA32><!WA32><!WA32><a href="http://www.mcs.anl.gov/mpi/index.html">MPI</a>,even attending one of the MPI Forums in Dallas. I gave a <!WA33><!WA33><!WA33><!WA33><!WA33><!WA33><!WA33><!WA33><!WA33><!WA33><ahref="http://www.cs.washington.edu/homes/grant/talks/llnl-mpi-class.ps.Z">tutorial on MPI</a> at LLNL, andlater <!WA34><!WA34><!WA34><!WA34><!WA34><!WA34><!WA34><!WA34><!WA34><!WA34><a href="http://www.cs.washington.edu/homes/grant/talks/boeing-mpi.ps.Z">another</a> at Boeing ComputingServices.<!WA35><!WA35><!WA35><!WA35><!WA35><!WA35><!WA35><!WA35><!WA35><!WA35><a href="http://www.purdue.edu"><!WA36><!WA36><!WA36><!WA36><!WA36><!WA36><!WA36><!WA36><!WA36><!WA36><img border=0 align=right src="http://www.cs.washington.edu/homes/grant/bkg-kid.gif"></a><p>The climate model I worked on was a parallel global climate modelcalled the <!WA37><!WA37><!WA37><!WA37><!WA37><!WA37><!WA37><!WA37><!WA37><!WA37><a href="http://www.cs.washington.edu/homes/grant/csmg.html">Earth System Model (ESM)</a>. The ESMis a framework for supporting parallel global climate modeling. Therequirements of the ESM project are that the model be transportable,scalable, and modular. The transportability of the code was deemed tobe crucial, as this would allow the group to take advantage of thelatest, most powerful multicomputers available. I ported the code toMPI, instrumented the code for post-mortem performance evaluationusing <!WA38><!WA38><!WA38><!WA38><!WA38><!WA38><!WA38><!WA38><!WA38><!WA38><ahref="http://netlib2.cs.utk.edu/paragraph/index.html">ParaGraph</a>,<!WA39><!WA39><!WA39><!WA39><!WA39><!WA39><!WA39><!WA39><!WA39><!WA39><a href="http://www.mcs.anl.gov/home/lusk/upshot/">Upshot</a>, <!WA40><!WA40><!WA40><!WA40><!WA40><!WA40><!WA40><!WA40><!WA40><!WA40><ahref="http://bugle.cs.uiuc.edu/Pablo.html">Pablo</a>, or some of myown tools, and worked on building a library of portablecollective-communication routines.<h3>Purdue</h3><p> I got my undergraduate degree in <!WA41><!WA41><!WA41><!WA41><!WA41><!WA41><!WA41><!WA41><!WA41><!WA41><ahref="http://www.cs.purdue.edu">computer science</a> and <!WA42><!WA42><!WA42><!WA42><!WA42><!WA42><!WA42><!WA42><!WA42><!WA42><ahref="http://www.math.purdue.edu">mathematics</a> at <!WA43><!WA43><!WA43><!WA43><!WA43><!WA43><!WA43><!WA43><!WA43><!WA43><ahref="http://www.purdue.edu">Purdue University</a>. My senior year Iwas president of the <!WA44><!WA44><!WA44><!WA44><!WA44><!WA44><!WA44><!WA44><!WA44><!WA44><ahref="http://www.cs.purdue.edu/acm/acm.html">Purdue Student Chapter ofthe ACM</a>. I also helped start a <!WA45><!WA45><!WA45><!WA45><!WA45><!WA45><!WA45><!WA45><!WA45><!WA45><ahref="http://www.cs.purdue.edu/upe/upe.html">UPE chapter</a>, andserved as the chapter's second president. Additionally, I helped toresurrect the defunct <!WA46><!WA46><!WA46><!WA46><!WA46><!WA46><!WA46><!WA46><!WA46><!WA46><ahref="http://www.math.purdue.edu/~mathclub/">math club</a> by servingas its treasurer, then vice president, and by helping to make it astudent chapter of the <!WA47><!WA47><!WA47><!WA47><!WA47><!WA47><!WA47><!WA47><!WA47><!WA47><a href="http://www.maa.org">MathematicalAssociation of America</a>.<br><br clear=all><p><center><small><em>Department of Computer Science and Engineering</em><br><em>University of Washington</em><br><em>Box 352350</em><br><em>Seattle, WA 98195-2350</em><br><em>Office: Guggenheim Annex 115 (<!WA48><!WA48><!WA48><!WA48><!WA48><!WA48><!WA48><!WA48><!WA48><!WA48><ahref="http://www.cs.washington.edu/homes/map/chateau.html">TheChateau</a>)</em><br><em>Phone: (206) 616-1854</em> <br><em>Fax: (206) 543-2969</em> <br><br>At school, I use a <!WA49><!WA49><!WA49><!WA49><!WA49><!WA49><!WA49><!WA49><!WA49><!WA49><a href="http://www.sgi.com">SGI Indy</a>.</small><br><!WA50><!WA50><!WA50><!WA50><!WA50><!WA50><!WA50><!WA50><!WA50><!WA50><a href="http://www.sgi.com"><!WA51><!WA51><!WA51><!WA51><!WA51><!WA51><!WA51><!WA51><!WA51><!WA51><img border=0 src="http://www.cs.washington.edu/homes/grant/sgi.gif"></a></center><h5><!WA52><!WA52><!WA52><!WA52><!WA52><!WA52><!WA52><!WA52><!WA52><!WA52><img width=100% height=5 border=0 src="http://www.cs.washington.edu/homes/grant/dot-green.gif"><br><!WA53><!WA53><!WA53><!WA53><!WA53><!WA53><!WA53><!WA53><!WA53><!WA53><a href="http://www.cs.washington.edu/homes/grant/"><!WA54><!WA54><!WA54><!WA54><!WA54><!WA54><!WA54><!WA54><!WA54><!WA54><img align=right src="http://www.cs.washington.edu/homes/grant/back-green.gif" border=0></a>Last updated August 30, 1996.<br><!WA55><!WA55><!WA55><!WA55><!WA55><!WA55><!WA55><!WA55><!WA55><!WA55><a href="mailto:grant@cs.washington.edu">Brian Kris Grant (<em>grant@cs.washington.edu</em>)</a></h5></body></html>⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?