[citation][nom]stifle[/nom]Those using server farms for simulation running generally write their own code / Linux kernels don't they?[/citation]
Not really, sure they might patch the kernel to accept some bleeding edge hardware but other then that the defacto standard in HPC is Perceus which is open source and by that nature free of charge. I find it hard to believe (unless you don't have linux admins) that free software that is very easy to use if you have been in the linux environment for 5+ years and are comfortable with linux is more expensive then a product that costs money. Would love to see the number magic they are using to back up this statement. If you don't have anybody in your organization that is comfortable in the linux environment you can purchase support from the Perceus team to help you set up and maintain your HPC clusters so maybe that is where they are trying to pose its cheaper over five years?
@Wheat_Thins, dude linux is just as easy to use as Windows.
I guess the real deal of linux in supercomputers is that IT'S COMPLETELY CUSTOMIZABLE. Don't like the kernel ? Change it. Because, lets be honest, there are no mainstream supercomputers, so you just can't write a kernel that performs 100% on all of the machines. That's why there's linux, an os which you can modify and shape to your needs.
Linux still will hold the majority. First off, all the companies running Linux will have been familiarized with it, and won't think it's difficult to use at all (which is really isn't if you don't mind sitting down for a couple hours and learning).
Secondly, as nevertell was saying, Linux is ENTIRELY customizable, and isn't under a million licenses and restrictions. You can make it do whatever you want with a little work.
I bet they got that being cheaper figure when factoring in support costs. As in microsoft will offer free support for a minimum of 5 years where as the linux guy would be inhouse and after customizations to the kernel, might be the only person capable or maintaining it without retraining making employee retention a high priority which can lead to higher overall salaries.
is Mr. Jobs doing consulting work for microsoft now? this sounds like an apple tax to me since linux is open source and even specific customized vendor server versions cost about 5000 percent less to purchase.
Works with CUDA, so now anyone with basic MS IT skills can build a "supercomputer" probably in a few days, right now.
eg: You have a couple of machines with some CUDA capable devices: You chain them together with the 180 day trial.
Bingo, you are now a potential scientist.
On thing they don't make clear is why the head node is running on a separate machine then the compute nodes? Wouldn't you just virtualize the head node onto a compute node box so it could be used for computing cycles as well?
Note that the HARC guys (or whatever they are called today) over at MSFT provide an unmatched support offering. When you're moving heavy iron or shooting for five nines these are the guys to have in your corner.
There is an engineer somewhere right now sitting at home playing xbox with his bag packed next to him. When a customer system goes down they'll work with someone on the phone while this guy gets on a plane.
[citation][nom]bfstev[/nom]I bet they got that being cheaper figure when factoring in support costs. As in microsoft will offer free support for a minimum of 5 years where as the linux guy would be inhouse and after customizations to the kernel, might be the only person capable or maintaining it without retraining making employee retention a high priority which can lead to higher overall salaries.[/citation]
That would be the key. Once a linux kernel is altered that much on a super computer platform, the resources to manage that system becomes limited and the employees wants more and more money and there is no way to turn him down.
On the MS side, the code is standard and you can always pick up a new employee to replace one with a big head.
The thing about building high performance clusters with Windows is:
(1) No one wants to waste time screwing around trying to understand the various licensing conditions.
With Linux, I download a distro and use it on as many machines as I want; real or virtual. If I want commercial support, I'll find a reputable party and pay for their services. If I choose to do things DIY, I can save money and learn something. (I often go for the latter because I love to learn new skills that could be put to practical use.)...Regardless, I'm in control of MY systems. I decide what's best for ME.
(2) The money you saved in NOT investing in Windows for your HPC implementation can be diverted for more crunching nodes (hardware).
(3) The inherit advantage of using Linux is the ability to modify the software (at a source code level), to meet your needs in a precise manner.
HPC goes with Linux like a glove because it plays to Linux's fundamental advantages.
Microsoft trying to fit Windows into markets like HPC and tablets doesn't work well because it doesn't play to Windows's own strengths. Its analogous to spending millions on trying to fit a square peg into a round whole...MS needs to seriously modify Windows until it doesn't even look like Windows; so as to fit into those markets. (Something it won't really want to do, as it costs time and money. Not to mention the need to come up with a different business model for it to be profitable.)
"Furthermore, Microsoft claims that Windows HPC Server is 32 percent to 51 percent less expensive than Linux-based HPC systems over five years."
Microsoft makes claims all the time. Its never realistic when you experience things in the real world. (Its lies to make their solutions look good).
I learned back when MS initiated their "Get the Facts" campaign (server comparison against Linux), they tend to intentionally create studies that result in their products looking better.
Thanks to Microsoft's marketing department, I've grown to be skeptical (always questioning) studies conducted in order to promote products or ideals/beliefs outside of the computing realm. So I guess MS isn't all bad! They've given me awareness!
Then again, I question the use of Microsoft solutions in mission critical applications...Just look at the following example.
The London Stock Exchange once used MS-based solution for about 4 years. It cost LSE US$65 million to build and implement the platform...The problem was, it wasn't reliable throughout its 4 yrs of service.
The final straw that broke the camel's back was on Sept 2008; The platform suffered a major outage that lasted for 7 hours. (This is right when the US Govt was rescuing Freddie Mac and Fannie Mae! Imagine not being able to trade during that period?!)...This also affected the Johannesburg Stock Exchange, as they use LSE's system!
You have to understand that 7 hours is several lifetimes; when you realise trading is seen from the millisecond perspective!
This was intolerable for LSE, as their previous system was running 6 years straight without a single outage! The following 12-months after the outage incident, the LSE gradually ripped out the MS-based platform and replaced it with a *nix based solution. (Consisting of Linux and Solaris with an Oracle back-end database).
Result? They saved approx US$15 million compared to the MS-based solution. (Due to reduced hardware requirements, licensing costs, and technical staff)...Not to mention they will save another US$17 million per year (for 2010 to 2012) because the *nix based solution is 6 times faster in trades.
In the LSE's case, not using Microsoft based solutions for really important things actually saved them money. Heck, it turns out the savings pays for itself in a much shorter time!
I find that a common theme in general. Switching to Linux is painful in the short term, but saves you quite a bit in the long term...The good news for gamers is that there is a way to implement the use of DirectX 10/11 support. (Not the way we've seen in Wine via DirectX-to-OpenGL translator. But via a new graphics stack that is currently underdevelopment.)
[citation][nom]twu[/nom]LOL,Marcus Yam's article has 217 words or 1412 letters. Aussie_Bear's comment has 700 words or 4246 letters.Aussie_Bear=Win[/citation]
You took the time to count every letter and word?
Article - Maybe took him for an hour or two to search/write.
Aussie_Bea - Typing speed 100 words/sec = 7 minutes + 2 minutes for his/her knowledge inputs.
Me - Word counts 10 seconds, 30 seconds to past it.
[citation][nom]twu[/nom]Article - Maybe took him for an hour or two to search/write.Aussie_Bea - Typing speed 100 words/sec = 7 minutes + 2 minutes for his/her knowledge inputs.Me - Word counts 10 seconds, 30 seconds to past it.[/citation]
How did you get 100 words/sec with 700 words = 7 minutes?
Okay so the $500.00 or whatever this will cost could have gone to half a decent server box running linux.
The add-on software to run your programs and the patches / upadtes costs how much + getting new techies to fix or maintain this new bugssoft on this untested distro that would obviously have holes / bugs and other issues.
All this when you already have a linux team running free software + software add-on's that is free and fixed weekly at slowest and not fixed by spending $300+ per upgrade / new release.
I wonder what the real cost would be over 5 years, man hours, down time on both and making a change from linux to Windblows.
If MS stated this can be used for home user / small office "5 humans in office server" HPC's or as a gaming rig with customizable schemes / layout's / themes to get it fast XP classic look / feel fast then sure I would buy what they sell with this statement.
PS> are they giving this HPC version away for free as that would be the only way to compete with linux / pelican HPC / other Linux HPC OS's.
There is nothing like a Linux custom source build system to make your day.