The collected works of SiliconBunny

SiliconBunny
Who joined April 12, 2005, 1:32 p.m.

Wrote on the subject of Itaniums? sgi to use Dorado line to boost workstations? at April 18, 2005, 3:04 p.m...

hamei wrote: Why would anyone want to buy this ? There's not a single thing about it that's special. You can buy anything you want from Sun and they haven't changed their minds every fifteen minutes about what their future is.

Because MIPS just isn't fast enough, no matter how balanced a system it's in.
Because making MIPS faster requires far more money now than it did in the past - the bar is much higher. Sun have quaterly revenue in the *billions*, and they can't do it anymore.
Because Sun don't have anything to touch this in terms of bandwidth.

Speaking of Sun, they have changed their roadmap massively:
- SPARC from desktop to enterprise, full binary compatability
- oops, we'll sell Intel kit as well now
- no Solaris x86 for you! Linux on our x86 kit
- oops, you can have Solaris x86 now. Sorry.
- behold our new dual core CPU!
- oops, no more UltraSPARC for you.
- behold our new multi-core horizontally scaling CPUs!
- meet Fujitsu, who will make our UltraSPARC replacements
- sorry, it's AMD64 at the low end and on the desktop now

Exactly what sort of roadmap is that?
Sun are now where SGI where 5-6 years ago - not enough revenue to continue to pay for extensive R&D, new chips getting more and more expensive to build (even with massively more volume than SGI at their peak), and introducing x86 stuff because it's what people want to buy.

hamei wrote: AND you can still run Solaris executables even on their newest workstations !

I'd like to see you running Solaris/SPARC executables on an Opteron workstation. Sun lost binary compatability across their range many years ago.

At least SGI are doing something about that. Sun just look embarassed and change the subject when you bring that up with them.

hamei wrote: Nothing on the desktop that SGI has done newer than the Octane shows any imagination whatsoever - so why spend four times as much money for half as much computer ? It was different when you were comparing an Octane to a Clunker, but hell - what has SGI done that's technically advanced or exciting since 1996 ? Nothing , as far as I can see.

Then you're not looking hard enough. The Origin 3000 is pretty advanced.
So's getting that architecture to work with Itanium.

I'd love to be a fly on the wall when you tell John Mashey that that stuff isn't exciting or technologically advanced.

Or is the market so flooded with scalable NUMA systems that Origin and Altix are now mundane?

Go and work with F25ks and you'll realise how far behind the curve Sun are.

CXFS is vastly more capable than any other shared filesystem out there. Do you have any idea how incredibly difficult it is to do something like that?

In what way is Onyx4 not advanced? Do you have that many machines with 32 graphics cores - that you can combine and split at will - that the Onyx4 is a mediocre box?

How many other machines have the bandwidth and raw CPU power of the Prism? Can you see any other machines - on the desktop - than can handle 4K imaging? In real time?

Have you seen the sort of bandwidth and sustained throughput that clusters of Altix and Origins can get talking to SGI storage? I fail to see how that is mundane or not advanced - it blows other vendors away.

There's more to life than the desktop.

In order to have any sort of R&D - or, indeed, any sort of future - SGI has to build and sell what the market wants.

This isn't 2000 any more. Companies don't buy IT equipment just because it's new and shiny. SGI are still innovating and still doing clever stuff, and still surviving because of that. If they were still doing pure MIPS/IRIX they would have gone under several years ago.

Would NASA have even bothered to talk to SGI for Columbia if it was Origin 3000 based? No, of course not - they would have gone to IBM.

I still look forward to the new machines from SGI, because they're still exciting, innovative, and ultimately very clever pieces of kit.

Cheers,
TOM

--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 1:32 p.m.

Wrote on the subject of Itaniums? sgi to use Dorado line to boost workstations? at April 18, 2005, 4:07 p.m...

foetz wrote: very good report, tom

*bows*

Cheers,
TOM

--

"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 12:32 p.m.
and authored 15 notes

Wrote on the subject of Itaniums? sgi to use Dorado line to boost workstations? at April 19, 2005, 12:11 a.m...

LaLora wrote:

Quote:

Would NASA have even bothered to talk to SGI for Columbia if it was Origin 3000 based? No, of course not

In case you still don't know - "Columbia" was almost entirely "donated" to NASA by SGI. They rushed to set it up quickly right before November Top 500 HPC list, so they could be on or near the top of the list for Altix "promotion" purpose.

I'm well aware of it - in what way does that change the fact that Columbia is still very technically advanced? Or that SGI would not have landed the deal if it was O3k instead of Altix?

I'd be very surprised if any of the bidders submitting quotes for Columbia asked for even 50% list price.

Once you leave the mass market toys and get to the big iron (16 way and upwards) no-one, ever, pays list price. At all.

'Headliner' systems like Blue Gene/L and Columbia are always *heavily* subsidised by the vendor. TOP500 is bragging rights, nothing else.

However, SGI could have given the kit away - for free - if it was MIPS based, and NASA still wouldn't have gone for it.

There's no loyalty in IT, and no-ones going to install a system that's significantly slower than the others on offer - especially not for a major installation like that.

LaLora wrote:

"Columbia", various sources estimate, at that time represnted around 10% of ALL global Itanium2 processor sales. When somebody has to "donate" 10% of all global sales of a new processor just to convince customers about their new system then that is more a sign of weakness for that company than some big success.

You appear to be confusing SGI with a mass market player like HP, Sun, or IBM.

They are not. They never have been, and they never will be. They are a niche player. They rule that niche, but it's still a niche.

Itanium is a niche CPU - anyone who believed it would be a mass-market solution can come to talk to me - I have a bridge to sell them. The market for Itanium is tiny - just as the market for MIPS or vector based systems is tiny.

The Earth Simulator was more than 10% of global vector-based systems sales when it was installed - doesn't mean it wasn't technically impressive, or that NEC were giving away kit out of desperation to stir the market.

High end system sales are very different from bulk desktop or 4 way server sales. Trying to quote numbers or systems shipped is just a nonsense in that space - that market doesn't work like that.

LaLora wrote:

Quote:

No, of course not - they would have gone to IBM.

It's quite funny that you wrote this line because:

"..If Itanium2 failes, SGI will be bought by IBM and their engineers will be used to boost new IBM systems with their experince.."

..was (or similar) written just a few months ago at HPC forum by someone close to IBM or SGI. Then, about a week later, I checked again at that forum and saw a very wierd thing - the posted message, and even all messages posted under (actually almost all messages on forum) were deleted (it's quite empty now), by administrator(s). Lots of people from industry come to visit TOP 500 site, sometimes even SGI has ads there, and SGI probably demanded for those messages quoting that they'll be bought by IBM be deleted.
It probably is not a secret in industry and business world any more that SGI will be bought by IBM in the end.

You don't honestly believe that? I mean, that's some serious tin-foil-hat paranoia going on there.

A post in a web forum does not a business decision make. Definitely not when it would involve hundreds of millions of dollars.

IBM will not buy SGI. At all. It's just not going to happen. Quite apart from the fact that SGI is dwarfed by IBM - an utterly insignificant player - IBM already have products in every area that SGI do.

Why on earth would IBM spend money to buy a non-competitive niche player when they already have higher volume sales from competing products?

Do you think IBM sales lose sleep at night when thinking of Altix vs. their POWER5/Linux business?

There's no business justification for it at all.

And that's not even touching on all the regulatory pain IBM would incur from the US goverment. Look at the hopes SGI had to jump through when buying Cray.

Governments don't like to see their technology suppliers being bought - it doesn't give them the warm fuzzies. They'd much rather bail them out with lucrative research grants and large scale projects. Look at how badly HP have screwed government departments over with their laughable Alpha 'road-map'.

The only people who might have had an interest in buying SGI would have been Sun - they need the high-end scalability. But not even Sun were willing to do a complete about face and swallow Itanium as their server CPU - so they jumped into bed with Fujitsu.

LaLora wrote:

The primary force of any high-tech company is always (and ONLY) innovation, new technology, helping new scientific research and so on. Once you loose that - it's over. You can't reinvent another HP zx6000 or claim to be serious scientific supercomputing company while posting news on your website that should attract religion-decision based shareholders and customers!

Which was what my post was all about.

SGI are still innovating, they're still ahead of the pack, and they therefore still have their little niche cornered. The day SGI stop innovating, you'll know about it, because they'll go under.

In the meantime, their stuff is technically excellent. The fact that it's no longer MIPS/IRIX should not blind any of you to their continuing innovation and - most importantly - their continued survival in a very harsh market.

Cheers,
TOM

_________________
--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 12:32 p.m.
and authored 15 notes

Wrote on the subject of Itaniums? sgi to use Dorado line to boost workstations? at April 19, 2005, 5:51 a.m...

schleusel wrote:

Prism still uses AGP 8x too. One can only guess why this is the case.

Alexis Cousein did some posts in the filth that is comp.sys.sgi detailing why this was.
It basically boiled down to the long development lead time for stuff like this (ie. AGP 8x was there and usable, PCI-E was a glint in some working group's eye) as well as stable and well-tuned chipsets surrounding the AGP implementation of the ATI cores.
I believe there was also some development issues/costs associated with the XBOW->PCI->AGP bridges - as in AGP is just more crap on top of PCI, whereas PCI-E is fundamentally quite different.

Cheers,
TOM

_________________
--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 12:32 p.m.
and authored 15 notes

Wrote on the subject of Prism workstation is out! at April 26, 2005, 10:37 a.m...

unixmuseum wrote:

More shots, see how it dwarfes the 23" LCD... For some reason, they're pushing CFD visualization quite a bit!

I was recently up at the BAR HQ near Silverstone. They have a lot of nice kit for CFD work. They were saying that the Onyx 3s they have didn't give them the performance they needed.

The CFD crowd will be big customers for this sort of kit, especially given the sort of insane bandwidth it can handle.

Cheers,
TOM

_________________
--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 12:32 p.m.
and authored 15 notes

Wrote on the subject of Prism workstation is out! at April 26, 2005, 1:56 p.m...

LaLora wrote:

..maby I better ask: What Linux do you get on CDs and installed on disk with SGI Itanium2 system purchase? Is it Novell (with all those GNU packages and OpenMotif) or some kind of SGIs own Linux distribution?

RHEL or SUSE. I believe you can choose which one. In theory you get the ProPack so you can install whatever you want, but there are support issues.

Cheers,
TOM

_________________
--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"

SiliconBunny
Who joined April 12, 2005, 1:32 p.m.

Wrote on the subject of Itaniums? sgi to use Dorado line to boost workstations? at April 28, 2005, 12:53 a.m...

Hamei, if you want to disillusion yourself into thinking that SGI are intellectually bankrupt, doomed, whatever - be my guest. I'm not going to argue with you.

But there are a lot of good people at SGI, doing some very clever things - don't bad mouth the company, and belittle those people's work, just because SGI isn't doing what you, personally, think they should be doing.

If you don't understand what SGI are doing, and why, maybe it's not because an entire company of clever people are wrong, and you are right - maybe it's because you are no longer their target customer?

Maybe you should run your own tech company for 4-5 years before making such sweeping statements about SGI's direction and achievements.

Cheers,
TOM

--
"Tell them we are not Gods, but SysAdmins, which is the next best thing,"