[LON-CAPA-admin] system overload

[Lucas, Mark]

On Sep 3, 2015, at 10:20 PM, Bob Gonzales <rgonzal at binghamton.edu> wrote:

> - test was comprised of simple problems, not published problems
> 
> Looking at differences from last year, 2 things stand out: the
> software raid 1 and the simple problems instead of published problems?
> Are simple problems cached on the access servers?
> 

Someone else can probably speak to this better, but the simple problems may
be at least part of the issue. The problem is really a template. I never really looked
at them very carefully, but the ‘content’ of the problem is essentially parameter information
that is stored somewhere, rather than a self-contained published problem that is 
cached. The template would be cached, but there may be a whole lot more information
that has to move back and forth to the library server than if these were regular published
problems.

Does someone else have ‘performance numbers’ or thoughts on the simple problems?

Mark

-- 
Mark Lucas 								email: lucasm at ohiou.edu
252D Clippinger Lab						phone: (740)597-2984
Department of Physics and Astronomy		fax: (740)593-0433
Ohio University
Athens, OH 45701