Re: Are Sunblade 1000s slow?
My experiances with dual and quad x86 systems that should blow an E4500 out
of the water have shown me that in multiprocessor applications, as the load
increases, the E4500 handles the loading much better.
My personal opinion as to why is related to the CPU cache and memory
architecture. In x86, each CPU has a smaller amount of cache and then
bottlenecks at the memory. In the E4500, even though each CPU is slower, the
higher number of CPU's and the layout of the memory makes it much harder to
slow down.
Now I need to get my hands on a good Quad Opteron to see how that feels!
On Tuesday 07 October 2003 17:52, Shawn Boyette wrote:
> On Tue, Oct 07, 2003 at 11:17:11PM +0200, Michael Andersen wrote:
> > Don't get me wrong, I am pleased to work with Sun hardware daily and I
> > use it privately as well - and although the design is more pleasing than
> > x86, there is no reason to pretend that the SPARC architecture is the
> > fastest today or in the future.
>
> This isn't a comment to Michael directly, I just want to vent, since
> we've gotten onto this topic.
>
> I've had this conversation over and over with x86-centric friends, who
> simply cannot understand that "fast" is not the be-all end-all of
> computing. SPARCs have never been about fast; they are about
> load-handling. The SPARC was made for, and grew up in, a multiuser
> environment where hundreds or thousands of processes are running at
> once. The x86 grew up in single-user machines where the priority has
> always been to do whatever the user is looking at as fast as possible.
>
> I do agree, though, that Sun's is just hurting itself by charging
> completely ridiculous prices for machines which are *basically*
> equivalent to what are considered "high end" servers in the x86
> world. The V100 costs $1k, and that's fine, but the v210 costs almost
> $6k if you actually want 2 CPUs. That's bad. A mostly-comparable dual
> G4 XServe can be had for $2.8k.
>
> I guess, were I an analyst, which I'm not, my analysis would be that
> Sun either needs to treat the low-end market like a low-end market or
> leave it.
--
S.W.
Reply to: