[Bioclusters] FPGA in bioinformatics clusters (again?) Ref: Cray and SGI

[Tim Cutts]

On 19 Jan 2006, at 6:48 pm, George Magklaras wrote:

>
>
> Tim Cutts wrote:
>> On 14 Jan 2006, at 5:07 am, Farul Mohd. Ghazali wrote:
>> SGI are in the same situation as  Cray; the US government will  
>> keep them alive as long as they need  support and no-one else has  
>> a competing product.
>
> SGI's position being the same as Cray's? You have made the  
> 'CrayLink' :-). After all I think Cray was briefly owned by SGI?  
> From a 'huge budget to even think about purchasing' point of view  
> maybe...But from a product point of view not so much under the  
> surface. The ALTIX maybe a nice package, it might contain  
> "customized NUMAlinks", but their technology is more on the  
> backplane and interconnect rather than the processor itself

Absolutely - see what I have to say in a moment.

> (my bias:personally I was never convinced about SGI offerings,  
> despite the fact I ran Origins a while ago). Cray might have  
> Opteron optimized systems (XD1) but their vector processors are a  
> different workhorse. If you now ask the question "how many people  
> need or can afford VP technology", the answer is almost certainly  
> very few. But my point is that SGI might be facing some tough times  
> due to the fact that they are trying to pass technology not so  
> unique as "unique", whereas Cray is trying to give you technology  
> that most folks do not need or cannot afford...
>
>>  This isn't actually  that bad a position to be in.  It's the same  
>> reason that VMS will  never die, and that HP are being forced to  
>> support Alpha for many years.
>>
> Another example of over-priced technology way ahead of its time...  
> and eventually, you got the Itanium with the "acceptable" compilers  
> some time after its first appearance :-p ...

The processor isn't what gives SGI a niche, as you say; you can buy  
less outrageously expensive Itanics from elsewhere, if you wish.   
What SGI have that is, at present, more or less unique, is the  
ability to build machines where you can scale CPUs, memory and I/O  
independently, and which you can scale essentially as far as you like  
until your money runs out.  This can be very important for some niche  
applications.

You want a machine with only 4 CPUs but a terabyte of RAM?  You can  
buy an Altix in that configuration.  I don't know of anyone else  
evenly vaguely mainstream that offers machines with as much physical  
memory in a single system image, although please enlighten me if  
there is one.  We have an Altix here with 4 CPUs and 192GB of RAM -  
not nearly as big as it could be, but still a configuration  
impossible to buy from anyone else, at the time (and still is, as far  
as I know).  I'd rather it wasn't IA64, but it's the memory that's  
important to us, not the CPU technology.  And it's the memory that  
makes them expensive anyway.  48 x 4GB DIMMs are going to be  
expensive regardless of whether you buy them from SGI or anyone  
else.  That RAM alone is around $50,000 at current prices (http:// 
osnews.pricegrabber.com/search_getprod.php/masterid=634351)

The NSA and their friends do have the big budgets you mention, and  
they want to search large databases as fast as possible.  That means  
holding the whole database in memory.  You also need to load that  
database into memory initially very quickly; the Altix can win there  
too - you just keep adding HBAs and spindles until the aggregate  
bandwidth reaches your requirements.  We've heard of examples where  
SGI have built systems to load terabyte databases into RAM from  
storage in a couple of seconds.

This is a market no-one else is really playing in, and because the  
likes of the NSA and GCHQ like this sort of machine, they're not  
going to let the one company go under, that can provide them.

That's my logic.  If free market economics was allowed to have its  
way, everyone would be right and SGI would collapse almost  
immediately.  But they have powerful friends.

Tim

PS.  I realise this is bioclusters, not comp.sys.sgi.advocacy, so I  
apologise for the off topic post - if anyone wants to continue this  
particular discussion we should take it off list.

-- 
Dr Tim Cutts
Informatics Systems Group, Wellcome Trust Sanger Institute
GPG: 1024D/E3134233 FE3D 6C73 BBD6 726A A3F5  860B 3CDD 3F56 E313 4233