📄 http:^^www.cs.wisc.edu^~pubs^faculty-info^vernon.html
字号:
Date: Thu, 07 Nov 1996 19:08:05 GMTServer: NCSA/1.5Content-type: text/htmlLast-modified: Tue, 31 Oct 1995 00:10:14 GMTContent-length: 2686<HTML><HEAD><TITLE> Home Page of Mary K. Vernon </TITLE></HEAD><BODY><H1> <!WA0><IMG ALIGN=MIDDLE SRC="http://www.cs.wisc.edu/~pubs/faculty-info/vernon.gif"> Mary K. Vernon </H1><BLOCKQUOTE> Professor of Computer Sciences and Industrial Engineering <BR> <BR> Computer Sciences Department <BR> University of Wisconsin <BR> 1210 W. Dayton St. <BR> Madison, WI 53706-1685 <BR> <BR> telephone: (608) 262-1204 <BR> fax: (608) 262-9777 <BR> email: <!WA1><A HREF="mailto:vernon@cs.wisc.edu"> vernon@cs.wisc.edu</A> <BR></BLOCKQUOTE><EM>Ph.D., University of California, Los Angeles, 1983</EM> <BR><EM>Interests:</EM>Techniques and applications of computer systems performance analysis,performance of parallel systems, parallel architectures and operatingsystems. <P><HR><H2> Research Summary </H2>I am interested in analytic modeling techniques and their applicationto computer system performance issues, with an emphasis on paralleland distributed system design issues. The modeling techniquesI have developed together with graduate students and colleaguesinclude the <EM>Generalized Timed Petri Net</EM>, and <EM>customizedmean value analysis</EM>. The GTPN can represent parallel systemfeatures such as synchronization and priority service precisely.The customized MVA technique yields a set of intuitive equationsthat represent parallel system features more approximately butcan be solved more efficiently. We have also recently proposedthe use of a technique called <EM>interpolation approximations</EM>for analysis of parallel processor allocation policies. This techniquemay also have broader applicability for parallel systems performanceanalysis. <P>Current research projects include characterization for high-performanceparallel workloads, analysis of parallel processor schedulingpolicies, scheduling issues in multimedia servers, and memorymanagement policies for networks of workstations. <P><H2> Sample Recent Publications </H2>The fairness of DQDB networks with slot reuse (with G. Brewster),<EM>Proceedings of the IEEE INFOCOM Conference</EM>, August 1995.<P> Accurate modeling of the hybrid hash join algorithm (with J. Pateland M. Carey), <EM>Proceedings of the ACM SIGMETRICS Conference</EM>,June 1994. <P> Use of application characteristics and limited preemption forrun-to-completion parallel processor scheduling policies (withS. Chiang and R. Mansharamani), <EM>Proceedings of the ACM SIGMETRICSConference</EM>, June 1994. <P> <HR><ADDRESS> This page was automatically created October 30, 1995.<BR> Email <!WA2><A HREF="mailto:pubs@cs.wisc.edu">pubs@cs.wisc.edu</A>to report errors.</ADDRESS><HR></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 + -