Minutes of the SIPB Meeting of 7.7.03 The meeting was called to order at 7:30 by golem. In attendance were Voting members: rax golem jenselby kenta merolish | glasser jtu ike Associate members: jhawk hartmans alex_c ocschwar | srz Prospectives: mwp eirl Guests: Treasurer's Report: hartmans: The treasurer is cool. He gave me money without me asking. golem: Was it legitimate? hartmans: Yes. Chairman's Report: Office Report: kenta: There are two broken things. Some RAM that we found didn't work is sitting in the bookshelves next to hodge. It is labeled Bad RAM. golem: You're sure it's bad? kenta: Yes. golem: Are we going to auction it? kenta: It might be able to be replaced. The pingtel phone has no dial tone and it is plugged in. ike: Ummm. Oh, umm. I'm going to be cleaning the office again tonight. I will take pictures with the digital camera per recommendation of chad. Office Cleaning Report: MIT Computing Report: kenta: Back in Aug 2002 I joined an MIT IS Discovery Committee on computationally intensive computing. In April 2003 we completed the discovery project. You may remember me asking about what MIT should do. The input I put in mentioned SIPB in exactly one way: If there is demand for a general-purpose compute cluster, SIPB may be interested in system administrating it. There is quite a bit of thought behind the conditional because discussion indicated that there may not be sufficient demand do support such a thing. I guess the hilights of the discovery project are that everybody and their mother has a Beowulf cluster and they tend to be more expensive to administer than people think due to air conditioning and power and system administration costs - grad students do system administration poorly. The committee wants to centralize some of the Beowulf clusters on campus. The prototype is somewhere on West Campus. The web page is web/is/discovery/intensive/ . hartmans: If this wasn't mentioned last week, IS paid people a lot of money for their webpage. ocschwar: A guy stopped by the office a few days ago to try to use unused cycles on cluster machines. I told him that there were political issues making this never happen. He should be by in a few days to become a prospective, and I don't know his name. kenta: It got closer than you can imagine. We discussed this and we almost made it to the end that IS should consider grid computing on Athena spare cycles. It was mostly cut because IS can't pay someone to write the code. I don't think the political problems are that big. The code is also half-written. ocschwar: I will point him at you. jhawk: I question whether this detail is appropriate to the meeting, but there is no one on the discovery team from Athena. (I have the list here.) The estimates you have may be wrong. golem: I have a question for Sam. Is that the webpage made by 7-year-olds? hartmans: No, web/is. golem: I was wondering how much money the 8-year-olds saw. Computer Services: jhawk: Charon has been acting up due to disk problems. SIPB Backup Report: Publications Report: jenselby: I suck and haven't done iMatlab yet but maybe I will soon. Orientation Report: IAP Report: CokeComm Report: kenta: What's up with the cups of pennies? jhawk: I think they have nothing to do with CokeComm. kenta: I move to donate them to CokeComm. golem: How about the office poker chip pool? kenta: I move it. jhawk: I think it is silly to have a motion for this. hartmans: I also agree that having a motion is bad, I think that this is a fine use for the pennies provided it doesn't live in the office. kenta: I withdraw the motion. Other: Other Other: golem: I bought a "Learn Mandarin in N days" book and am probably going to fail to learn Mandarin but want to be able to read some street signs in Hong Kong and China between August 11th and September 10th. glasser: Will the meetings be in Mandarin? golem: The meetings will be in English. The meeting was adjourned at 7:41. Minutes taken and submitted by rax.