Hi,
Can anyone help me understand if I have understood the performance calculations correctly for the report service:
Current configuration settings for the report services:Queue time limit of the report service (seconds) = 360
Non-peak periodNumber of
high affinity connections for the report service during non-peak period = 2
Number of
low affinity connections for the report service during non-peak period = 8
Maximum number of processes for the report service during non-peak period = 6
Peak periodNumber of
high affinity connections for the report service during peak period = 2
Number of
low affinity connections for the report service during peak period = 8
Maximum number of processes for the report service during peak period = 6
My calculations of the total report service performance:4 report service instances x 6 report service processes = 24 total number of report service processes on all dispatcher servers
24x2 = 48 total number of high affinity connections for the report service during non-peak period
24x8 = 192 total number of low affinity connections for the report service during non-peak period
24x2 = 48 total number of high affinity connections for the report service during peak period
24x8 = 192 total number of low affinity connections for the report service during peak period
Questions1.) Does this mean that I'm able to serve interactive users with totally 192 reports executing in parallel?
2.) When I running 100 reports in parallel as interactive user I don't understand the figures below.
This is how the current process usage looks like when running 100 reports in parallel as interactive user:
ReportService instance #1 - Number of processes = 2/6
ReportService instance #2 - Number of processes = 2/6
ReportService instance #3 - Number of processes = 2/6
ReportService instance #4 - Number of processes = 1/6
3.)
We get some reports failing initially when we suddenly start running 100 reports in parallel as interactive user...
Is it because the total number of report service processes isn't ready and need to start in order to serve all requests?
Does "Queue time limit of the report service" at 360 seconds mean that reports will fail if they exceed 360 seconds during start up of report service processes I mentioned previous question?
4.) Is it possible to define the following configuration setting in Cognos?
- Minimum number of processes for the report service during non-peak period = 3
- Minimum number of processes for the report service during peak period = 3
I haven't found any similar setting, but maybe there is some kind of workaround to achieve this?