[samigo-team] Samigo performance--limits and solutions

[Jim Mezzanotte]

Hi David,

Thanks for the info, very helpful. I wasn't aware of the performance
degradation as students progress, so that's good to know.

In regards to setup, et cetera:

db is mySql
load distributed across 3 servers, with a fourth utility server for SIS

Organization of assessments might also be impacting
performance--assessments consist of numerous sections, and each
section has just one question drawing randomly from a pool (pools
average 20-60 questions).

I did try reorganizing, so there were fewer sections with more
questions, but that didn't seem to make any appreciable difference.
Still, given what you're saying about question pools and degradation
as students progress, this might be a perfect storm of conditions, all
impacting performance.

Best,
Jim Mezzanotte
rSmart



On Wed, Apr 10, 2013 at 4:05 AM, David Horwitz <david.horwitz at uct.ac.za> wrote:
> Hi Jim,
>
> Can you tell us more about your setup (version, database) & the assessment
> organisation? These all have an impact on performance and how it effect you.
>
> Some observations from us. We run exams in the sizes you mention - our
> cluster is sized the way it is largely to accommodate this.
>
> There are 2 distinct load issues we have seen:
>
> - starting an quiz is expensive, particularly if the questions are drawn
> from a pool. Rutgers identified an issue that's fixed in 2.9.0 and Sam's
> fixes should further improve matters here. We advise the staff giving large
> exams like this to try stagger starts to help mitigate this.
>
> - The performance on moving to next degrades as the student moves to the
> next question. I did some work on this under
> https://jira.sakaiproject.org/browse/SAM-1523 that i thought was in trunk
> but i see isn't.
>
> D
>
>
> On Tue, 2013-04-09 at 14:19 -0400, Bryan Holladay wrote:
>
> https://jira.sakaiproject.org/browse/SAM-1695
>
>
>
> Subtask 2 & 3 are verified in production.  Subtask 1 is still waiting for
> David to reply about production statistics.
>
>
>
> 2 & 3 have shown a 2000% reduction in the number of DB calls
>
>
>
> On Tue, Apr 9, 2013 at 2:07 PM, Jim Mezzanotte <jmezzanotte at rsmart.com>
> wrote:
>
> Hi all,
>
> We're encountering Samigo performance issues with large (800+) course
> sites, and specifically in cases where groups of students (up to 200)
> are taking an assessment simultaneously in one physical location with
> the Respondus lockdown browser. There's not a way to decrease this
> concurrent usage, so I wanted to get feedback from others on your
> experience with Samigo performance limits, and any steps you've taken
> to maximize performance.
>
> I'm referring specifically to speed of navigation between
> questions/parts for students, as well as speed of access for
> instructors scoring an assessment. Another thing I'm trying to
> determine is whether access time for concurrent usage changes
> according to these parameters:
>
> a) usage distributed across the tool in multiple sites
> b) usage for multiple assessments in a single site
> c) usage for a single assessment in a single site
>
> Determining this poses a challenge with manual testing (I'm looking
> into automated testing). But thanks in advance for sharing any
> experiences/solutions/best practices.
>
> Best,
> Jim Mezzanotte
> rSmart
> _______________________________________________
> samigo-team mailing list
> samigo-team at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/samigo-team
>
>
>
> _______________________________________________
> samigo-team mailing list
> samigo-team at collab.sakaiproject.org
> http://collab.sakaiproject.org/mailman/listinfo/samigo-team
>
>
> ________________________________
> UNIVERSITY OF CAPE TOWN
>
> This e-mail is subject to the UCT ICT policies and e-mail disclaimer
> published on our website at
> http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27
> 21 650 9111. This e-mail is intended only for the person(s) to whom it is
> addressed. If the e-mail has reached you in error, please notify the author.
> If you are not the intended recipient of the e-mail you may not use,
> disclose, copy, redirect or print the content. If this e-mail is not related
> to the business of UCT it is sent by the sender in the sender's individual
> capacity.