Recent Posts

Pages: 1 2 3 [4] 5 6 ... 10
31
SERT / Re: SERT 2.0.6 Workload Efficiency score (Storage)
« Last post by Takumi Takahashi on July 10, 2023, 10:43:25 PM »
Thank you for your reply.

I checked and found that the disk write cache was disabled.

Please let me know if there are any other items I should check.

Also, is this storage score within the normal range? We would appreciate it if you could share any data that would be helpful.



32
Hello Joseph,

Sorry for the problem.  Can you try to run with logging enabled, and forward the logfile?  See Section 3.1.3 in the SERT User Guide on where the logfile is located, and attach those files for analysis.
33
Hi pgalizia

Thank you for your reply
I've confirmed that nodes can pin each other and the firewall is already disabled.
still not resolves the issue.
34
SERT / Re: SERT 2.0.6 Workload Efficiency score (Storage)
« Last post by pgalizia on July 06, 2023, 10:16:48 AM »
Hello Takahashi-san,

Your Storage worklet score will vary, depending on the number of drives used, the types of drives (HDD, SDD, M.2, etc), and where they're located on the bus.

Also, be sure to disable your disk write caches as per section 4.1.2 of the SERT User Guide.
35
SERT / SERT 2.0.6 Workload Efficiency score (Storage)
« Last post by Takumi Takahashi on July 06, 2023, 05:07:42 AM »
In SERT 2.0.6, the storage score exceeded 300 when measured with the following OS and CPU combinations.

Is this result reasonable?

Please let me know if there is any missing information.

Windows Server 2019 / i3 -12100TE
Windows Server 2019 / Celeron G6900TE
Windows Server 2022 / Pentium Gold G7400TE

Best regards,


36
Make benchmarks and methodologies transparent, enabling users to replicate results and gain a clear understanding of the testing process. Document the benchmark setup, configuration, and measurement methodologies comprehensively. Continuously assess the relevance of benchmarks to emerging technologies, such as cloud computing, virtualization, AI/ML, and IoT. Regularly update benchmark suites to incorporate new technologies, frameworks, and hardware architectures. Foster a collaborative environment within the SPEC community to encourage feedback, suggestions, and contributions from users, researchers, and industry experts. Regularly engage in discussions, conferences, and workshops to gather insights and address evolving requirements.
37
Hello Joseph,

This error usually occurs when there's a connectivity issue between the devices.
  • Can you ping one device from the other?
  • Have you tried disabling your firewall before starting SERT?
Give these two items a try and see if that resolves your issue.
38
Thanks for the reply.

I adjusted the value of "cpuwarmupIntervalCount", and it improved.

39
SERT / SERT2.0.6 Unable to establish status connection with director on port 8897
« Last post by Joseph on July 04, 2023, 12:38:50 PM »
Hi SERT Team, when I using the SERT 2.0.6 to test the multi-node system(128 cores per CPU and 2CPU per Node, total 6 nodes ), I encountered the following issues.

"Unable to establish status connection with director on port 8897
 Q: Is Uilistener enabled in listeners.xml on local director?
 Timed out waiting for test suit to cancel.
 Restarting the GUI may be necessary to re-launch the test successfully....."

I've tried to edit the director-run script (director.bat/director.sh) to Xmx1024m, Xmx40964m...etc, but still does not work.

Please let me know if there is any workaround, thank you
40
Hello Takahashi-san,

Please see Section 2.4.6 of the "Server Efficiency Rating Tool (SERT) Run and Reporting Rules" PDF.

The CPU worklets perform three warmup iterations before the calibration/measurement iterations.  You can edit the "config-all.xml" file on the UI side to modify the number of warmup iterations run to reduce the run-to-run variability:

Code: [Select]
<interval-count id="cpuWarmupIntervalCount">3</interval-count>
Try adjusting that value higher or lower, and see if that improves your LU variability.
Pages: 1 2 3 [4] 5 6 ... 10