Welcome to my personal home page.
Here are some other sites with info about me:
- UCSD CSE homepage
- This page has my research info and publications list.
- Weblog
- Flickr photos
- del.icio.us links
AboutWelcome to my personal home page.Here are some other sites with info about me:
By admin at Oct 5 2006 - 11:33 | login to post comments
An exhaustive list of my projects and technical experienceAs an addendum to my resumé, here is an The list is very roughly in order of amount of experience. (NOTE) This is currently incomplete. By Mike at Dec 19 2007 - 12:30 | read more
Stakeout with Growl SupportIn my old blog, whose archives I haven’t updated yet, I wrote about my Stakeout program, which watches for changes in files using kqueue and runs a script you give it as an argument. I wrote a second version called ‘gstakeout’ that will then also try to send a growl notification to inform you of the results of the script it ran. The links to those programs fell off the net a while ago, but they are still available in this tarball: stakeout-2.tgz As of this writing, I haven’t tested it in years, but I would be surprised if it didn’t work. Feel free to let me know if anything’s wrong, especially if you fix it yourself! AddressService - An OS X System Service that searches your AddressBook entriesAddressService is a system service I wrote to automate the task of finding someone’s address or phone number to copy and paste into an email or chat message. The service works on your current selection - just select a part of someone’s first or last name and select either “Address/Insert Addresses” or “Address/Insert Phone Numbers”. Your AddressBook is searched for matches, and the selection is replaced by the address or phone number for all the matches. All addresses and phone number variants are inserted, and you just delete all the ones you don’t want. There is a universal binary version of the service here: AddressService-Universal What's Working in HPCJournal Paper : What's Working in HPCSource:Cyberinfrastructure Technology Watch Quarterly, Volume 2, Number 4 (2006)Keywords:HPC, productivity, queueAbstract:Productivity in High Performance Computing ("HPC") systems can be difficult to define, complicated by the sometimes competing motivations of the people involved. For example, scheduling policies at many centers are geared toward maximizing system utilization, while users are motivated only by the desire to produce scientific results. Neither of these motivating forces directly relates to the common metric widely put forward as a measure of merit in HPC: high code performance as measured in floating-point operations per second (FLOPS). This paper evaluates some factors contributing to the net gain or loss of productivity for users on today's HPC systems, and explores whether or not those factors are accurately being accounted for in the way systems are evaluated and scheduled. Usage patterns are identified through job logs and ticket analysis, and further explained with user surveys and interviews. This paper reveals insight into productivity on current HPC systems, where users' time is spent, what bottlenecks are experienced, and the resulting implications for HPC system design, use and administration. Notes:This paper is part of the Darpa HPCS program - the "P" in "HPCS" denotes productivity.Export:By admin at Oct 5 2006 - 12:33 | 1 comment
|
Navigation |