[Bioclusters] quick look see at fractal computing.

[James Cuff]

On Wed, 22 Feb 2006, Kathleen wrote:

> We have limited the query size initially in order to manage a surge in
> usage, which we experienced today.  If you want to really blast us, please
> contact Nick Robertson at nick at massivelyparallel.com.  He'll get ya hooked
> up so you can test our system with a massive query.  You can also talk with
> one of our mega users who blasts us at least once a quarter.

Cool.  It would be nice to see some real figures.  I had guessed my test 
that only showed a fairly paltry 2.6x speed up compared against a badly 
configured single server was not quite right.  Along with only 62% of the 
default number of hits being returned I knew there must have been a 
mistake...

I'm sure than in an ideal world your approach ought to totally smoke a 
single server on the end of a damp piece of ethernet to a fairly heavily 
laden NFS server?

Am I right in thinking that there are what, 250+ servers on your backend 
farm no?  I guess not all of them are dedicated to the portal mind you.

I did notice that my little test job started instantly, so you must either 
have the work queue really well managed or I was lucky at the time and it 
was totally empty.  Good to hear that you had a surge in usage today,, you 
got my attention at any rate.

As I mentioned, I'm sure that the vendor agnostic teams (of which there 
are quite a few) that hang out on this list would love to give your method 
a good solid workout.

That's if you are up for some compare and contrast.  On the other hand 
there could be some real good marketing in there at the end ;-)

I'm certainly not one of those chaps I'm afraid, but it would be good to 
see some real numbers as they come out, it could end up being very useful.

I just wanted to get the ball rolling really, I'll duck out of the 
thread for now.

Best,

j.