http://parlab.eecs.berkeley.edu/wiki/ 2009-11-24T08:12:41-08:00 http://parlab.eecs.berkeley.edu/wiki/ http://parlab.eecs.berkeley.edu/wiki/lib/images/favicon.ico text/html 2009-09-09T11:48:32-08:00 http://parlab.eecs.berkeley.edu/wiki/bootcamp09?rev=1252522112 2009 Par Lab Boot Camp - Short Course on Parallel Programming Title Presenter(s) videowmvslides Day One -- Wednesday, August 19 Introduction and Welcome Dave Patterson (UCB)stream[wmv]pdf Short Course on Parallel Computing Jim Demmel (UCB) stream[wmv] pdf Introduction to Parallel Architectures Part 1 John Kubiatowicz (UCB) stream[wmv] pdf Introduction to Parallel Architectures Part 2 John Kubiatowicz (UCB) stream[wmv] pdf Shared Memory Programming with Threads Katherine Yelick (UCB & LBNL)strea… text/html 2008-02-13T15:03:00-08:00 http://parlab.eecs.berkeley.edu/wiki/calendar?rev=1202943780 <http://www.google.com/calendar/embed?src=i0fca6dr7a6sfb8gph6ujuca4o%40group.calendar.google.com&ctz=America/Los_Angeles> text/html 2008-11-30T15:10:16-08:00 http://parlab.eecs.berkeley.edu/wiki/carazvan?rev=1228086616 Posters Presenter(s) Students Other(s) Title Brief abstract [pdf] [ppt] Razvan CarbunescuRazvan Carbunescu, Meriem Ben Salah, Mehrzad Tartibi Phil Colella, Panos Papadopoulos, Tony Keaveny, James Demmel, Katherine Yelick ParLab Health Application We provide a desciption of the physical problem being modelled and the computational tools used in the process and their success in previous problems text/html 2008-10-16T13:44:56-08:00 http://parlab.eecs.berkeley.edu/wiki/coffee?rev=1224189896 This is a place to discuss the coffee/coffee maker in the Par Lab space. Give us thoughts, ideas, what you like, what you hate, any other things relevant to coffee needs. text/html 2008-02-26T17:30:14-08:00 http://parlab.eecs.berkeley.edu/wiki/filedrop?rev=1204075814 File Drop text/html 2008-03-19T13:15:20-08:00 http://parlab.eecs.berkeley.edu/wiki/foobar?rev=1205957720 this is a test text/html 2008-01-31T15:11:25-08:00 http://parlab.eecs.berkeley.edu/wiki/glossary?rev=1201821085 Moved to an internal page. text/html 2008-04-10T10:13:18-08:00 http://parlab.eecs.berkeley.edu/wiki/iw?rev=1207847598 Mission: Understand market shifts/transitions/trends in our industry to identify new and viable business opportunities needing our research/technology/expertise 5-10 years in the future. More concretely, the goal is to identify new products, services, or business models to apply our research and expertise, especially in markets that are not directly pursued by large players (meaning they will not be directly competing with our ideas for some time). text/html 2008-03-27T16:43:33-08:00 http://parlab.eecs.berkeley.edu/wiki/iw_discussions?rev=1206661413 Upcoming Events April 1-3, 2008: Santa Clara convention center <http://www.multicore-expo.com/index.php> 2008-03-19: 4pm 545R * Summary of Jim Smith's talk from 3/13 (notes on wiki) * Startups related to parallel computing (see links and articles below) * Summary of meeting with Kurt * Brainstorm/Propose new startup ideas we have been thinking about this past week. text/html 2008-03-27T10:41:25-08:00 http://parlab.eecs.berkeley.edu/wiki/iw_meeting_notes?rev=1206639685 2008-03-19: Meeting Notes If you don't have time to read the entire email, please read 1) Action items, and 3) Future Meetings. I am especially interested in getting feedback on my proposal for structuring future meetings (section 3). 1) Action Items: text/html 2008-04-01T00:41:58-08:00 http://parlab.eecs.berkeley.edu/wiki/iw_taxonomy?rev=1207035718 Parallel Computing Markets vs. Solution Providers Matrix We will use the following matrix to guide our meeting discussions. Markets listed have a wide variety of problems related to parallel computing. Companies listed are solutions providers that address parallel computing problems. Place a '+' in the box if a company provides solutions for a particular market. text/html 2008-03-28T09:12:59-08:00 http://parlab.eecs.berkeley.edu/wiki/market_for_finance?rev=1206720779 Computational finance requires significant processing power to churn through complex economic models. Low latency transaction times are crucial, some estimates that a 1ms advantage over competitors means $100M a year in revenue(1). There is significant parallelism embedded in underlying algorithms, where most algorithms are based on PDE solvers, Monte Carlo simulation, simplex or interior point methods (Matrix Algebra). text/html 2008-11-26T11:08:10-08:00 http://parlab.eecs.berkeley.edu/wiki/paintball?rev=1227726490 text/html 2009-11-09T15:25:21-08:00 http://parlab.eecs.berkeley.edu/wiki/patterns?rev=1257809121 [Our Pattern Language] (Comments) [Data Parallel Design Patterns] The patterns: the desktop view Pattern Template: a checklist of what you should include in a pattern Pattern Workshops: Monthly meeting to discuss and improve patterns The contributors text/html 2008-06-16T12:36:58-08:00 http://parlab.eecs.berkeley.edu/wiki/related_work?rev=1213645018 * High-level Parallel Languages * Low-level Parallel Languages * Auto-Tuners * Scheduling * Memory Consistency text/html 2009-10-23T10:10:04-08:00 http://parlab.eecs.berkeley.edu/wiki/rules?rev=1256317804 Noise * Be considerate of noise level in the cubicle area. * Try to keep discussion quiet and of a technical nature. * If you are being bothered, ask the people to “get a room” rather than leave. * Longer conversations should move to the lounge area or kitchen. * If you receive or need to make a telephone call, use a meeting room or step outside the lab. text/html 2009-05-28T10:22:40-08:00 http://parlab.eecs.berkeley.edu/wiki/start?rev=1243531360 Welcome to U.C. Berkeley's Parallel Lab web and wiki page. Starting in February 2005, a multidisciplinary group of Berkeley researchers met weekly to discuss the implications of the parallel revolution for applications, programming models, architecture, and operating systems. These “Berkeley View” discussions created a synergistic environment that led to the creation of the Parallel Computing Laboratory (“Par Lab”) starting officially January 21, 2008. text/html 2008-10-24T20:57:52-08:00 http://parlab.eecs.berkeley.edu/wiki/upcrc_report?rev=1224907072 <date range covered by report> Prepared by <name> Executive Summary Section Overview: • Overall UPCRC status: Any project highlights/lowlights, admin details, disclosures, etc. • Visibility of UPCRC – things they have done to promote mainstream client parallel computing and engage with academia, industry, funding agencies, public policy makers, media Key Changes: