Ayende @ Rahien

It's a girl

UberProf performance challenges

One of the things that the profiler is supposed to do is to handle large amount of data, so I was pretty bummed to hear that users are running into issues when they try to use the profiler in high load scenarios, such as load tests.

Luckily, one of the things that the profiler can do is output to a file, which let me simulate what is going on on the customer site very easily.

The first performance problem is that the profiler isn’t processing things fast enough:

image

The second is that there are some things that just totally breaks my expectations

image

There are solutions for those problems, and I guess that I know what I’ll be doing in the near future :-)

Comments

Romain Verdier
12/14/2009 10:55 AM by
Romain Verdier

Is the rhino in the waiting dialog animated?

Glenn F. Henriksen
12/14/2009 11:10 AM by
Glenn F. Henriksen

So that's 53k queries in one data context? Seems a bit much :D

Ayende Rahien
12/14/2009 11:12 AM by
Ayende Rahien

Yes, that is NOT something that the profiler expects to see, I am afraid to say.

Glenn F. Henriksen
12/14/2009 11:18 AM by
Glenn F. Henriksen

What's with the loading picture, 160.062 out of 11.957? Textbox overflow?

Ayende Rahien
12/14/2009 11:23 AM by
Ayende Rahien

Glenn,

160K loaded, 11K remaining to be processed.

Glenn F. Henriksen
12/14/2009 11:29 AM by
Glenn F. Henriksen

Ah, I guess it's easier to see it when you see the numbers counting.

smurf
12/14/2009 02:29 PM by
smurf

if you write about features of ÜberProf, do they affect all profilers or do i need an ÜberProf license to get them?

Ayende Rahien
12/14/2009 04:50 PM by
Ayende Rahien

smurf,

When I am talking about UberProf, I am referring to all profilers.

Paulo Quicoli
12/14/2009 05:42 PM by
Paulo Quicoli

I dont know internals of your loading data... but, maybe, use some of parallelism from .NET 4 would be good. :)

John Farrell
12/14/2009 06:51 PM by
John Farrell

Bah, thats nothing, I've done 1.6 million in a single context before. :p

Why? It was quicker to drag a table into EF than write the sql query manually for a simple nightly processing job. Performance overhead for using an ORM was in "who cares" territory so it worked.

Comments have been closed on this topic.