Re: Are Sunblade 1000s slow?
David S. Miller wrote:
> I'd get an Opteron if I needed 4GB+ of ram.
Except that Sparc is far more widely supported for these kinds of
applications. Not everyone writes open source applications, unfortunately.
Still, much rather an Opteron than an Itanium.
> I in fact like Apple hardware, and I think Apple is viable because they
> make a computer that is very fun to use. They give something of
significant
> value to a large group of people. A very small (and ever shrinking) group
> of people see this when they look at what Sun is offering.
Fair point. Although Sun is working on changing the whole basis of its
existence, so it may be more a case of people moving away from the Sun
hardware model to the Sun software model. Although I really do not envy
Sun's management, they have some very tough choices to make.
> Let me put it this way, if you think 4-way is the limit of scaling for
> an x86 box what do you need a many-cpu system for? If it's database work
> get a bunch of 4-cpu x86 boxes and setup an oracle cluster or something
> similar. Clusting allows a ton of big end problems to be solved and
> nothing beats the cost effectiveness of a rack of tiny x86 blades.
Absolutely. Unfortunately many software vendors have not yet caught up with
the concept of clustering. It takes a lot of work and some serious
forethought to write an application that handles clustering properly.
Multiprocessor systems are a lot more forgiving of bad software
architectures. And believe me, there are some really bad candidates out
there.
(Which reminds of a favourite anecdote: a customer wondered why their four
processor box was not performing, so they stuck in another four processors.
No improvement. A five minute analysis revaled the application was single
threaded. Doh!)
> If you still absolutely need so many cpus, fine go get the box from
> Sun and be their support headache. As their market share shrinks,
> it's you the customer that is going to have to help Sun defray that
> situation by paying more and more to get their systems and services.
Well, Apple went through a tough time too. It is now Sun's turn. Maybe they
can alter their strategy to provide systems that are more relevant to
people's needs. Maybe they can pull off the utility computing model better
than IBM. Maybe Sun and Apple merge? Maybe IBM just buys Sun and moves
everyone onto MIPS?
Also, don't forget that the IT industry has always had huge inertia. "You
won't get fired for buying IBM" is still around, except that we now
s/IBM/Wintel/g. Luckily for Sun, the same attitude still applies to them as
well. Whether they can maintain this attitude or will fall by the wayside...
well only time will tell.
> I don't agree with this either. Why did we have that "race to the moon"
> the other year between AMD and Intel to get to 1GHZ? It's because the
> volumes pay for all the R&D much better than Sun's model does.
But the volumes also have much lower margin? Anyway, I think you are
confusing competition which can just as easily happen with high end systems
with commodification, which is the process of making something as cheap and
as widespread as possible.
> Who has the best silicon etching and processor design technology out
> there? It's not Sun and Texas Instruments, I can assure you of that :
I thought it was IBM? :)
> > Aw, can I have some. Please?!? <tries best doe eyed look>
>
> Sorry :)
Drat. ;)
> Well, not against Linux/Sparc since there's no version of Matlab
> you can get for that.
Good point.
- Matthew
Reply to: