Message boards :
SETI@home Science :
Change rattles the world's biggest dish
Message board moderation
Previous · 1 · 2
Author | Message |
---|---|
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
Really? My experience is that Windows is always slower than Unix. Not if you have the latest Intel CPU. Mine is only an Opteron 1210 at 1.8 GHz with two cores. But the CERN Virtual Machine takes only one core, leaving the other to other BOINC projects or to a second Virtual Machine running Solaris Express. On that I have installed a BOINC client and a SETI app by Dotsch and that is really slow, but only because the client or the app see my Opteron only as a pentium_pro+mmx, while it is SSE3 capable. |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30640 Credit: 53,134,872 RAC: 32 |
I think you misunderstood. Two boxes, identical except the O/S. Run the same program and the Windows box will take longer to finish. Depending on the mix of I/O this might be from hardly measurable to quite a bit. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
Even if this is true, today the Windows boxes could not run a Scientific Linux program without the CERN Virtual Machine. Now they are running it, at least in the Alpha test project, and they are far more numerous than Linux boxes or Macs, thus giving more chances to the scientists to have their data processed. Tullio |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30640 Credit: 53,134,872 RAC: 32 |
Even if this is true, today the Windows boxes could not run a Scientific Linux program without the CERN Virtual Machine. Now they are running it, at least in the Alpha test project, and they are far more numerous than Linux boxes or Macs, thus giving more chances to the scientists to have their data processed. That is Cern, not SKA. If you read the item SKA is going to build its own crunch farm. Why would they pick a slower O/S that then needs to run a virtual machine, slower still, for their work? They wouldn't. It isn't that hard to port code between different Unix flavors. There are some factors at work here. At SKA they have just one science program to support, where as Cern has to support thousands. That is the reason Cern took Red Hat and modified it for their use. Not everyone has jumped on their bandwagon, nor should they. If I read correctly SL barely will run open office as a lot of standard Linux is not included. SKA may need to run a much wider set of standard applications, things which SL may not support. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
When Test4Theory starts on my Linux box it opens only a terminal with no GUI and Scientific Linux programs run in it. No Office, it only runs scientific programs like Pythia, based on MonteCarlo methods. What I am saying is that most distributed computing volunteers use Windows, not Linux or Mac OS X. If SKA, like SetiQuest, seeks only Linux or Mac users. good luck to them. Tullio |
John Neale Send message Joined: 16 Mar 00 Posts: 634 Credit: 7,246,513 RAC: 9 |
Let us not forget the Square Kilometer Array project: A bit late into this thread, but... The problem with the article in this link is that it gives the impression that the SKA will be located in Australia and New Zealand. The fact is that site selection will take place in 2012, and Australia (with New Zealand) is competing with South Africa (with eight other southern African countries) for the right to host the SKA site. Here are some links: The official SKA website South Africa's SKA website Australia's SKA website From what I've read in various places, the Australian consortium is considering Skynet. And from what I've read here, Nereus runs under MS Windows. |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
You are right on both counts. I was led in error by s passage in a Computerworld article: Instead, the researchers prototyped a Nereus V testbed using 200 clients at the University of Western Australia, Curtin University and iVEC with two servers deployed through management at a NereusCloud domain. The clients used for processing included Mac Minis and some Linux-based desktop PCs with at least two gigabytes of memory. Tullio |
Gary Charpentier Send message Joined: 25 Dec 00 Posts: 30640 Credit: 53,134,872 RAC: 32 |
Something tells me the SKA isn't going to be using Boinc http://www.skatelescope.org/the-technology/signal-processing/ Approximately 160 Gigabits (10e9) per second of data will be transmitted from each radio dish to a central processor meaning that the dishes alone will produce ten times the current global internet traffic! The use of aperture array receptors will further increase data rates to many Petabits (10e15) per second which represents more than 100 times the current global internet traffic! |
tullio Send message Joined: 9 Apr 04 Posts: 8797 Credit: 2,930,782 RAC: 1 |
Probably SKA, if and when completed (where?), will need a supercomputing powerhouse more than some bunch of volunteers like BOINC. Maybe it will use some quantum computer, like the one recently sold by D-Wave to Lockheed Martin for ten million bucks. As an AQUA@home volunteer I have contributed to this project but I was not happy about this first sale to a weapons factory. Tullio |
Slavac Send message Joined: 27 Apr 11 Posts: 1932 Credit: 17,952,639 RAC: 0 |
Something tells me the SKA isn't going to be using Boinc Woah. Executive Director GPU Users Group Inc. - brad@gpuug.org |
©2024 University of California
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.