It's all about the answers!

Ask a question
QuestionsTagsUsersBadges
Guidelines | FAQ
Follow this question
Question details

×864
×33

question asked: Sep 03 '13, 12:50 p.m.
question was seen: 3,597 times
last updated: Sep 05 '13, 2:20 p.m.

Related questions

What are expected performance times for RSA DM operations?

0
Ricky Fielding (38●7●12) | asked Sep 03 '13, 12:50 p.m.

Hi All,

We are experiencing very long operations when using RSA Design Explorer to access our models on the DM server as well as some long operation times from the DM web interface. Compared with the RSADM 4.0.1 Performance and Sizing report, many of our numbers are twice the expected wait time, and some are much, much worse.

Client type Use case Mean response time
(seconds ± σ)
RSADM 4.0 RSADM 4.0.1 4.0.2 
50 users 50 users 100 users Our Experience
(single machine) (single machine) (distributed deployment)
Web client Expand an explorer tree node 1.95 ± 1.37 1.60 ± 0.50 2.54 ± 1.14 2-3 sec
Create a resource 6.38 ± 2.65 2.96 ± 0.96 5.13 ± 1.96 5-10 sec
Save a resource 5.86 ± 2.39 2.45 ± 0.79 4.32 ± 1.69 11 sec
Lock a resource 0.99 ± 2.12 0.20 ± 0.15 0.37 ± 0.57 1 sec
Unlock a resource 1.16 ± 0.84 0.20 ± 0.14 0.32 ± 0.25 1 sec
Open a resource 7.10 ± 2.99 3.35 ± 1.40 7.04 ± 2.18 12 sec
Open a diagram 1.55 ± 2.07 0.14 ± 0.14 0.26 ± 0.32 25 sec
Add a comment 0.82 ± 0.93 0.09 ± 0.09 0.13 ± 0.26 1-3 sec
Search all resources 0.80 ± 0.71 0.63 ± 0.35 0.95 ± 0.52 5-7 sec
Search diagrams 3.54 ± 3.37 1.52 ± 1.22 2.25 ± 1.69 1-3 sec
Get OSLC representation 0.96 ± 1.64 0.38 ± 0.26 0.56 ± 0.40  
Rich client Expand an explorer tree node 4.23 ± 1.17 0.32 ± 0.18 0.53 ± 0.47  4,15,45 sec
(3,15,50 items)
Open properties for a UML resource 0.56 ± 0.95 0.61 ± 0.74 1.08 ± 1.00 2-11 sec

The servers are hosted by another organization and we access the desktop applications through Citrix.  We have verified that the Citrix connection can almost double the response time, but even when using the tools local to the hosting organization, we had some very long response times.

In the performance and sizing report (https://jazz.net/library/article/1214) do we have any information about the model that was being used?  Response times for expanding a tree node seem to be proportional to the number elements at the next level.  For a small node with 2-5 elements, we experienced 2-5 second wait times.  But for a node with 50 items in it, we saw at least a 22 second delay (using the local connection) and as much as a 45 second delay when connecting externally.  Is there a chance that this is expected behavior?  We never have more than a handful of users on at one time, and it is likely (though I can't verify 100%) that most of our tests have been conducted with a single user accessing the server.

Unfortunately, I have not been given any insight into how the Jazz environment is organized.  I'm told that  recommendations in the sizing report have been followed.  Even if we can fix the longer delay times, our best case numbers so far may force us away from RSADM.

Thanks,

Ricky

One answer

Most liked answers ↑|Newest answers|Oldest answers

0
permanent link
Dusko Misic (196●1) | answered Sep 05 '13, 12:42 p.m.
FORUM MODERATOR / JAZZ DEVELOPER
The model used for sizing report is 110 MB on disk. The numbers you are seeing are not normal, so something else must be going on here. Which version of RSA DM are you using? Any chance you could get info on how the server is set up (for example, which app server, db, authentication, etc)? Another useful thing to diagnose the problem would be to get logs from the server around the time you see most delays.

I would also suggest opening a defect against DM and attaching the logs there.
Comments
Ricky Fielding commented Sep 05 '13, 2:20 p.m.

Thanks for the response. I'm hopeful our hosting organization will be willing to respond with answers to some of your questions, and, potentially, to open a defect.  I was mostly hoping other DM users could respond with similar information about delays.

My experience has been that delays in opening tree nodes are roughly proportional to the number of elements in the node. Should that be right?

Also,

1) Did the model in the IBM sizing report have large nodes?  Some of the nodes in our models contain 180 elements, which open 50 at at a time. 

2) Our server host is concerned that the report represented an extreme best case and isn't really a valid comparison for their setup.  Hopefully they'll be able to respond with a little more info on the comparison.

3) Is there any info on how the HTTPS requests from the rich client are structured and what kind of SQL queries those turn into?  I've wondered if there are single or multiple requests per element and whether that could be causing delays...perhaps some sort of limiter/governor on HTTPS or SQL connections? 

Your answer

Register or to post your answer.


Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.