Open Source SpeedShop Project Opened 14
drjzzz writes "Federal Computer Week reports that the National Nuclear Security Administration of the US Department of Energy is paying about $3 million of a $6.8 million collaboration between Silicon Graphics and the Universities of Maryland and Wisconsin to develop an
open-source version of SpeedShop, SGI's performance analysis tools. This will redress what a SGI engineer characterizes as scarce analysis software for Linux. A "Pro" version will also be developed and sold by SGI. Maybe even those of us without access to ASCI White can tweak our boxen to do 3D simulations of complete nuclear detonations, NNSA's main interest. Now that's what I call homeland security and real respect for the spirit of the second amendment."
A legit question (Score:3, Insightful)
I do have to admit that SGI is the only company that has made me interested in Itanium in any kind of way.
Re:A legit question (Score:2)
Actually we are one of the biggrst SGI customers, having over 2K of their SGI workstations, and over 10 clusters each at least 128 nodes.
I am sorry though i cant tell you the company name (wish i could), but SGI does have a wide customer base from where i stand, and i think they are making good money.
Re:A legit question (Score:2, Interesting)
Will your company stay on SGI gear after they kill off MIPS and Irix and go Linux/IA64? Or are you "solution shopping" for the future?
Re:A legit question (Score:2)
Re:A legit question (Score:1)
I bet most of their customers are looking forward to the time when they introduce Itanium workstations. According to SGI people, they have MIPS binary compatibility, and the old MIPS programs actually run faster on Itanium processors with a runtime bina
Re:A legit question (Score:3, Funny)
SecurIT Informatique Inc. Montréal, Québec?
(security is everyone's responsibility)
Distrobuted computing work (Score:1)
I once did a distrobuted computing project whom wouldn't allow a linux version of their software to be created due to the software people somewhere in the system having such control as that the code could not be shared by a lower company. (Bit weird but stick with me).
It wasn't nuclear anything, it was cancer research, ~(stick with me some more)~ but still it all comes down to lots of data to process in the end. Distrobuted computing projects do this extrely well (As we all kn
Lack of nuclears?! (Score:2, Interesting)
Much to most of the Slashdot communities pleasure, I am sure, there are not many nuclear engineering schools left [univsource.com].
This should be developed by one of the nuclear engineering firms, such as AREVA and not a bunch of students, regardless of whether it is to be open source or not (which most code in the nuclear field ends up being that is used like this).
Re:Lack of nuclears?! (Score:1)
Re:Lack of nuclears?! (Score:1)
SGI and University (Score:3, Interesting)
Too many times I hear of stories where students time and effort can't be capitalized. They work just end up counting toward school property.
funny, I would not have said this (Score:2)