24 posts • joined Thursday 25th October 2012 15:27 GMT
Re: Have we learned nothing....
"Now whats the value in knowing the other Eu ministers position on agricultural subsidy cuts or on whether Scotland would be allowed into the Eu without the Euro?"
No such thing as a discussion on EU agricultural cuts!
Re: And still
In any case, a SMP system should be able to scale better than a cluster running message passing. While sitting here writing out this reply, I can't think of any application that would scale better in an MPP system vs. SMP, but I could be wrong.
Sorry you are wrong. There is always an overhead, (cache coherency), which the cluster does not have, so scaling classical MPI code like CFD will "always" scale better than any/all ccNUMA systems. ScaleMP does well with CFD because inherently in HW it's a cluster. Where is doesn't do well is in mixed workloads, e.g. CFD and FEA structural analysis on the same system.
Re: Object: It's not just about storing stuff...
Have you seen what we are developing by any chance?
No obvioulsy no sod all about HPC, as they are almost all running Linux all ready.
Re: More like Windows 8 RTM than 8.1
Why are we surprised Win8 was crap. Every 1st version is shite and they fix it, (mostly), with the second version. This has happened since Windows 95.
If only someone would produce an Officesuite that was really inter-operable with Office, I could get away from Balmerland
Re: Still pugging something
USB = flash?
That will be same flash that EMC, NetApp, HP, IBM, Fusion-I/O, Violoin and Uncle Tom Cobely use in there enterprise arraye, tool
Re: This is journalism?
Why do we keep believeing the hype, despite time after time, they fib! e.g. XEON Phy, was going to compete with GPGPU without re-writing your code! The truth is you can run standard code on Phy and it runs like a dog, because the processing units have a much slower clock that standard XEONs. You get the benefits from Phy by using the vector units, which is a significant task for a good programmer, same as GPGPU.
Intel like Microsoft might just be a dinasaur 64.999Million s years ago!
Never discount Intel on t he grounds that can't "through horespower" at the problem. They havedone that for as long as I can remember, and look at what happened to:
Motorola 68K, Mips, and Sparc??
I dont think it mmatters at all. IBM's large HPC installations, are either PSeries or iDataplex if x86 based, which they are keeping hold of.
Re: not surprised
Perhaps we are assuming x86 has a future anyway, I'm not sure it does. Intel has managed to fab ist way up market for years, with a pretty poor architecture, but pretty soon they will see the limits of physics. On the other hand ARMs will be coming from Calxeda, via Dell, HP et al, and they will compete well versus x86 for web servuces. So IBM becoming an ARM licensee, flogging off the x86 business for $5B might well be a very clever move.
Thats coz you live in the sticks mate
Re: Smartphones are for yesterday's cool kids.
Ever thought of making friends instead?
Re: And this will cost...
Indeed. The problem with ALL flsah drives is teh sustained write performance falls off a cliff, once you exhaust the write cache on the drive. Because of this flash has very jittery performance too, do hardly ideal or high frequency trading..
1. 32Tbyte shared memory boxes are not common
2. Solaris is STILL a far more robust OS than Linux
3. SPARC is expensive versus x86, but if you have SMP needs, then revert to #1. above
Re: But, will it run Crysis?
How many bloody times the same "funny" post!
Re: M5-32 is not the only one with 32TB RAM
SGI have much larger, ccNUMA, systems than 32TB
Well they were called Bronto-burgers in the Flintstones. Probably didn't eat the head, so knew no better!
Increasing CO2 levels makes plants grow faster. Its what happens in a greenhouse, so shouldn't matter to the plants at all.
Re: I can't say
Oh dear I can feel the late night call of a Kerby sales pitch
Re: I can't say
Shite at sucking up dirt though