Gas comes from Norway
It's a minor point, but I believe most of our gas, unlike the rest of the EU, comes from NOrway and not Russia
The flame is the gas, not the author
106 publicly visible posts • joined 10 Feb 2009
"But anyway, for this kind of tasks, you don't need an interconnect, just buy a POWER 780 with the highest CPU clock available.."
For what kind of tasks? Most HPC tasks, require a balanced: CPU, interconnect, (for message passing), and I/O performance, so to say these codes just need a Power780 is pretty shortsighted. If you can get the fastest CPU then you can get a way with a less efficient interconnect. If your code is I/O bound then it doesn't matter how slow you CPU or interconnect are, if they are constantly in I/O wait.
As regards x86 and shared memory, memory addressing on XEONs maxes out at 16TB, and for something the NCSA is buying, I would be surprised if that was big enough.
Interesting that a Tier1 CTO, finally, tells it how it is and identifies some of the fundamental problems flash has. Add to this the appalling performance difference between 10% full and 90% full flash devices and you start to see it isn't the panacea that the flash suppliers claim, oh and neither will phase change either.
Links, to what?? Cluster computing is a group/cluster of individual machines, which are linked together with an interconnect, (a network). Interconnects are either:
Ethernet based for capacity computing, think particle physics work, or finance, e.g. QM or
Capability based, whereby you require very fast, low latency message passing between distributed codes. e.g. CFD, Computational Chemistry codes. Low latency interconnects are: Quadrics, SCI, Myrinet and Infiniband, plus a few proprietary ones from Cray, IBM< and this Fujitsu.
All compute bodes are standard linux boxes, running a fairly standard Linux OS, although you might have added specialist maths libraries. Hence why the scalable linux statement was silly.
If you still want links look up "High performance computing".
You suggest that all molecules which are derived from Benzene, e.g. Phenol, Toluene/Methylbenzene, TCP are carcinogenic, but only Benzene, with no additional molecular sub-group, is carcinogenic as far as I know. It's like saying oygen and water are very similar because they both have oxygen atoms present?
Why would a manufacturer not build a tablet with Android, if the market wants it? I think your faith in MSTs power is a little out of date, unless you are Nokia of course!
They might build any because the ARM is a better solution for Android on a tablet, but if the market exists "they will come".
Most HPC is in finance where many small single threaded Monte Carlo simulations are running. This would be followed by semiconductor guys who also have mostly serial codes, Oil&Gas, the same, followed by Gov labs and Universities who, yes run the Grand Challenge benchmarks like Linpack, (which is an utterly pointless exercise), but most University work is probably serial or maybe 4 way parallel.
It always made sense to me that Cisco, (because of the commoditisation of x86 HW, would eventually become a serious systems company. I thought an agreement with NetApp made perfect sense. NetApp = EMC so it still makes perfect sense.
If I was HP, who have been turned into a really big Dell, by Hurd, should be worried.
Morten,
You are comparing apples with oranges. You compare an IBM system, with a new technology, which I assume you haven't tested?, with an SGI system, which I assume you have?
Give these people a break, at least until you have had your fingers in the keyboard and can quantity you claims.
What I can't get my head around is the girl makes a claim of sexual harrassment one day, then says she never menat Hurd any harm the next and this isn't picked up by anyone?
Also how greedy can you be when you earna basic of $15M and fiddle your expenses. I am waiting for the bit where Sid James and Kenneth Williams join the HP board, (if they were still alive)!
The only area, (I can think of), where Windows HPC cluster makes sense, is in mechanical CAD. where the engineer starts on a Winodws box, with a big graphics card, and then run sout of power. He has no interest in learning a new OS, (at any level suh as linux), as he is an engineer, and so HPC cluster for him makes sense. A better solutuion would be a Windows scalable SMP, but they don't exist YET!
Verari, RackGI are not big enough tin shifters to compete with very large tin shifters, e.g. Dell and HP, and even IBM. They had an angle for a couple years when the big boys stayed loyal to Intel, and they grabbed market share as AMD had a better CPU. That changed 3 years ago, and since then they have been on borrowed time. I would love to meet the VCs who put ~$50million into Verari, smile and ask WTF were you thinking of?
It's a sad state of affairs, but being cheap isn't enough any more and innovating is getting harder and harder to do.
Are you sure the xbox 360 is a Power box? Also Power7 is not a follow on to cell, it's a standard processor versus a highly specific parallel engine.
More cores is only good if you can use them, so the point about needing parallel development tools is valid. You also need a shed load of parallel devlopers and they don't exist.
Another way of solving the problem of efficient use of multicore is a product called MCOPt, which claims to manage resources better than the stanard linux OS, so you get more work through multi-ore without re-coding. I know some people in two V-large semi-conductor vendors who say it works very well. www.exludus.com
CHeers
Todd
Nothing said here shows Ellison as wrong.
1. HW business is crap, e.g. HP and Dell = low margins
2. Orcale sw business = Vhigh margins
3. Oracle needs HW vendors less than they need Oracle
4. Sun HW, good database platforms
5. Databases need good storage I/O and backup, =Sun
6. Sunacle sounds like a good business!