Re: [Beowulf] Re: Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Mark Hahn
while I like the idea of these being available, I wonder where the real (big) market is. You mean other than commercial / HR databases that were built on Sun's SMPs and now have a questionable upgrade path? when I look at Sun's high-end boxes, I see either webservers (the many-thread stuff) or

Re: [Beowulf] Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Franz Marini
On Thu, 2009-05-28 at 15:04 -0400, Mark Hahn wrote: > > Given that JuRoPa (the new cluster at the Juelich Supercomputing Centre > > (JSC) : http://www.fz-juelich.de/jsc/juropa ) is scheduled to be in > > production by the end of June, I would say that is more than probable > > that at least for se

Re: [Beowulf] Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Mark Hahn
2010 is not so far away anymore, and Nehalem-EX will probably be there before Tukwila anyways. It feels like this announcement is a way for Intel to make the Itanium branch less and less appealing. ia64 has been dead for years. SGI merely provided the embalming fluid that kept the corpse from s

Re: [Beowulf] Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Franz Marini
On Thu, 2009-05-28 at 16:45 +0200, Kilian CAVALOTTI wrote: > > nehalem-ex sounds exciting, but it kind of spoils the effect if it > > really won't hit the streets till sometime in 2010. > > 2010 is not so far away anymore, and Nehalem-EX will probably be there before > Tukwila anyways. It feels

Re: [Beowulf] More casualties in the HPC landscape

2009-05-28 Thread Ashley Pittman
I can't speak for SiCortex but the Quadrics news as reported on the register is spot on, I'm surprised however that it's taken so long for the news to filter through into the main-stream. Ironically enough SiCortex were one of the first people I sent my CV to :( Ashley, On Thu, 2009-05-28 at 17

[Beowulf] syslog to sql

2009-05-28 Thread Michael Di Domenico
Does anyone have any opinions on which (free) programs work best for sucking syslog data into mysql? I had used syslog-ng in the past, but it looks like they went commercial on the sql import side. ___ Beowulf mailing list, Beowulf@beowulf.org sponsored

Re: [Beowulf] Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread John Hearns
2009/5/28 Kilian CAVALOTTI : >> > IIRC what our SGI sales rep told us, SGI will be working on a x86 version of > NUMAlink for their Altix series, using Nehalem-EXs instead of Itanium2s. So > customers in the SMP market whose only choice was Itanium will probably turn > away from it to adopt the x86

[Beowulf] More casualties in the HPC landscape

2009-05-28 Thread Kilian CAVALOTTI
Sad day today. It looks like Quadrics and SiCortex are closing doors: http://insidehpc.com/2009/05/27/inside-track-sicortex-rumored-to-be-closing- its-doors/ http://insidehpc.com/2009/05/27/quadrics-prepares-to-shut-its-doors/ On the bright side, Cray bought its debt back: http://insidehpc.com/20

Re: [Beowulf] Re: Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Joe Landman
Jason Riedy wrote: code, and the sequential/slow parts are faster. The main difference will be in I/O, but bumping the memory way up will make I/O less important for many uses. Er ... ah ... no. It will only delay the inevitable. It won't eliminate it. Blue Gene units have massive main m

Re: [Beowulf] Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Kilian CAVALOTTI
On Wednesday 27 May 2009 19:58:07 Mark Hahn wrote: > > Oh, and Machine Check Architecture (MCA) is coming up to Nehalem-EX, too. > > yeah, well, reg got that wrong. They got that from Intel: """ Nehalem-EX will add new reliability, availability and serviceability (RAS) features traditionally fo

[Beowulf] Re: Finally, a solution for the 64 core 4TB RAM market

2009-05-28 Thread Jason Riedy
And Mark Hahn writes: > while I like the idea of these being available, I wonder where the > real (big) market is. You mean other than commercial / HR databases that were built on Sun's SMPs and now have a questionable upgrade path? Likely replacing current mid-range, <100-node clusters with a si