who said flash mob computing is fizzling itself out? here's the latest status:
1) kishore makes a pragmatic suggestion: choose a location where we have existing dozens of pc with infrastructure like power, backup, ac, network, etc. we had volunteers here devoting their sites. kishore recommends use one of these sites to showcase a flash mob supercomputer. even if it is of 50 nodes, or 100 nodes, it is a proof-of-concept. perhaps in the second or third attempt we could break a national or other record. perhaps in the hundredth. kishore knows. once upon a time, he was a car rally freak. participating and finishing is important. 2) raj stumps everyone with his 21 unanswerable questions. solution: ignore his questions and get on with the flashmob computing, as per kishore's suggestion. a scaled-down, initial flash-mob recce is better than no flashmob computer. simple. some tests done by flashmobbers showed problems in the nodes actually 'subtracting' rather than 'adding' computing power. unfortunately, vivek khurana, on whose shoulders the history of this moment is to rest, was missing during this discussion. so we could have discussed alternatives. i believe we could attempt this with another live distro, or contact the flashmob supercomputing list to find out what is happening. please, let's have a list of volunteered sites for the flashmob event. let's do a sample test of this using bootable cds on-location. even a two-node cluster would do. vivek, you can best coordinate this, as you are completely clued in to the intiative, and tell us how do we proceed. :-) LL (flasher in the mob) _______________________________________________ ilugd mailinglist -- [EMAIL PROTECTED] http://frodo.hserus.net/mailman/listinfo/ilugd Archives at: http://news.gmane.org/gmane.user-groups.linux.delhi http://www.mail-archive.com/[EMAIL PROTECTED]/
