From owner-chemistry@ccl.net Sun Mar 26 09:46:01 2006 From: "Laurence Cuffe Laurence.Cuffe ~ ucd.ie" To: CCL Subject: CCL: Wireless cluster Message-Id: <-31330-060325195635-24532-TNnFDgQi2k9KpBXwgExo1A,,server.ccl.net> X-Original-From: Laurence Cuffe Content-disposition: inline Content-language: en Content-transfer-encoding: 7BIT Content-type: text/plain; charset=us-ascii Date: Sat, 25 Mar 2006 23:56:27 +0000 (GMT) MIME-version: 1.0 Sent to CCL by: Laurence Cuffe [Laurence.Cuffe_._ucd.ie] ----- Original Message ----- > From: "Andrew D. Fant fant-$-pobox.com" Date: Friday, March 24, 2006 7:32 pm Subject: CCL: Wireless cluster To: "Cuffe, Laurence " > Sent to CCL by: "Andrew D. Fant" [fant=-=pobox.com] > John McKelvey jmmckel%x%attglobal.net wrote: > > > > Please don't laugh, or maybe this is so funny [ridiculous?] that > you'll > > get a good laugh, and it will make your day! > > > > Anyway, I want to build a small linux cluster [6-8 total > processors] but > > don't have a lot of cooling in one place. Running wires would > not be > > practical, if not impossible Now, the application is extremely > > coarse-grain, and only a very, very small amount of data gets > moved > > about once initialization of a job is done.. Can it be done > "wireless?" As a more general problem cabling is a nightmare on big clusters, and I could see quite a niche marked for high bandwidth short haul wireless systems in the future. One other very odd thought on this was when I got my latest mobile phone, and realized it had the same amount of memory as the first computer I used for computational chemistry, with 3 times the processor speed, I wondered if you could build a grid computing cluster using either old mobiles, or the existing network. All the best Dr Laurence Cuffe > Hey John, > I wouldn't make it my first solution, but it could be done. As > others have > pointed out, your bandwidth would be limited, your latency would be > rathervariable, and you would be stuck in half-duplex mode for your > communications.You might want to look at more sophisticated methods > for data transmission to > make up for this if the process isn't truly embarrassingly > parallel. In > particular, coding with MPI even if it isn't strictly required for the > parallelism might pay dividends, since you would be able to use > gather/scatterand broadcast/reduce code to sling what data you have > around, which might be > more efficient than what you can do from the shell. Alternately, > if you break > down and get a good wireless router that supports multicast well, > you might want > to look at using multicast to handle communications. Again, the > idea being to > announce data that other systems might care about and let the ones > thatactually DO care grab it from one set of packets. Then again, > they are doing > wonders with pantone-matched drop cables and discrete baseboard > mounting clips > these days, too. ;-) > > Hope this helps, > Andy > > -- > Andrew Fant | And when the night is cloudy | This space to let > Molecular Geek | There is still a light |------------------ > ---- > fant- -pobox.com | That shines on me | Disclaimer: I > don'tBoston, MA | Shine until tomorrow, Let it be | even speak > for myself > > > > -= This is automatically added to each message by the mailing > script =- > To recover the email address of the author of the message, please > changethe strange characters on the top line to the \a/ sign. You can > also> Conferences: > http://server.ccl.net/chemistry/announcements/conferences/ > Search Messages: http://www.ccl.net/htdig (login: ccl, Password: > search)> > -+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+- +-+- > +-+-+ > > > >