Recent Posts

Pages: 1 [2] 3 4 ... 10
12
SPECvirt_sc2013 / SPECvirt_sc2013 benchmark results using performance metric
« Last post by genepetti on January 22, 2024, 06:28:18 PM »
Hello,
According to the FAQ on the SPEC website, SPECvirt_sc2013 is expressed as "SPECvirt_sc2013 @ <5*Number_of_Tiles + Number_of_DBservers> VMs".

As I understand, the right-hand side expression denotes the number of VMs utilized on the virtualized SUT (System Under Test), at peak performance.

However, examining for example the benchmark results for Hangzhou H3C Technologies Co., Ltd UIS R390, the score is SPECvirt_sc2013 638.6 @ 37 VMs.

The physical host has a processor with 32 threads, i.e. logical processors (16 cores x 2 threads per core). The VMs on the host are configured as follows: 1. Web Server VM: 2 vCPUs 2. App Server VM: 3 vCPUs 3. Mail Server VM: 1 vCPU 4. Batch Server VM: 1 vCPU 5. Infraserver VM: 1 vCPU

Total: 8 vCPUs

All of the above VMs are required for the execution of 1 tile. Furthermore, for each set of 4 tiles, a DB server VM is required, which is configured as follows: - DB Server VM: 6 vCPUs

According to the benchmark result, at peak performance 37 VMs are utilized, of which 2 are DM Server VMs, which means:

7 tiles * (Web Server VM + App Server VM + Mail Server VM + Batch Server VM + Infraserver VM) + 2 * DB Server VM

Hence 7 * 8 vCPUs + 2 * 6 vCPUs = 68 vCPUs are utilized at peak performance.

But considering that 1 vCPU = 1 logical processor on the physical host. How is it possible that 68 vCPUs are utilized when the capacity of the physical processor is only 32 logical processors?

Hope that you can explain this for me. Thanks
13
SPECvirt_sc2013 / Re: something wrong with appserver
« Last post by icklesassa on January 16, 2024, 02:06:50 AM »
thank you!
14
SPECvirt_sc2013 / Re: something wrong with appserver
« Last post by colemer82 on January 16, 2024, 01:41:21 AM »
I think these resources can be useful to you:

SPECvirt_sc2013 Documentation: https://www.spec.org/virt_sc2013/

15
SPECvirt_sc2013 / Re: something wrong with appserver
« Last post by TravisHindley on November 09, 2023, 11:30:22 AM »
Hi Bob,

Apologies for any trouble with the SPECvirt_sc2013 benchmark. It looks like there is some kind of version mismatch going on here. Could you please provide us with the details on your test harness? Specifically, what version of the kit are you using and did you apply any benchmark updates to either your prime controller, SUT VMs., or your clients.

Thanks!

Travis
16
SPECvirt_sc2013 / something wrong with appserver
« Last post by Bob on November 02, 2023, 08:30:37 PM »
2023-11-0216:23:26:254setting hostsReadytrue
2023-11-0216:23:26:254 PrimeControl:calling getBuildNumber()
2023-11-0216:23:26:255 PrimeControl:buiLdNum[2] =80
2023-11-02 16:23:26:255 PrimeControl:getting job id=2
2023-11-02 16:23:38:874 PrimeControl:buildNum[1]=80
2023-11-02 16:23:38:898 PrimeControl:getting job id=1
2023-11-02 16:23:38:908 specvirt:waiting on 1 prime clients
2023-11-02 16:25:41:829 PrimeControL:buildNum[0]= 79
2023-11-02 16:25:41:832_PrimeControl:getting job id=0
2023-11-02 16:25:41:835 PrimeControl:[ERROR]masters
  • build numbers(79) do not match the specvirt prime controllers(80)Please update complete harnessand retry

2023-11-02 16:25:41:835 PrimeControl:[ERROR]startMasters() failed!
2023-]-02 16:25:41:835PrimeControlsending abort Test  to prime clients
2023-11-02 16:25:41:836PrimeControl:setting job:6
2023-11-0216:25:41:836PrimeControl:got job : 6   id=0
2023-11-0216:25:41:836PrimeControl:switch:jobNumber=6;id=o
2023-11-0216:25:41:836PrimeControl:got job:
17
SERT / Re: SERT 2.0.6,2.0.7
« Last post by Mr.Brown on September 28, 2023, 07:39:20 AM »
Hi pgalizia,

sorry for the delay in this case.
I resolved both issues i had with the following solutions:

1. for the LU Worklet i tested all range numbers from 1-10 and in my case with number 2 the error message were gone.
2. i managed to clear this issue with the "Range Discovery" Test after that a "Full Test" was okay with no error.

Thank you for the help and for the recommendations.

Kind Regards
Mr. Brown
18
PTDaemon / Re: YOKOGAWA WT310E connection
« Last post by GregDarnell on September 21, 2023, 01:46:45 PM »
Hi Tien,

PTDaemon LXI support works well on Ubuntu 18-20. Later Ubuntu versions have a change in their RPC interface that breaks the third-party LXI code. I assume that same change affects RHEL as well, so older RHEL versions may also work.

Greg
19
PTDaemon / YOKOGAWA WT310E connection
« Last post by tienguye on September 21, 2023, 01:36:32 PM »
Hello, when I try running the following command
./ptd-linux-x86 -L 49 x.x.x.x

It gave me the following error on RHEL 9.2 OS
                               SPEC PTDaemon Tool
                        Version 1.10.0-ed9a21d2-20220817
                     Licensed Materials - Property of SPEC
     Copyright 2006-2022 Standard Performance Evaluation Corporation (SPEC)
                              All Rights Reserved.
  For use with benchmark products from SPEC and authorized organizations only.

Selected power meter 'Yokogawa WT310' from wt310.cpp
Calculated PTD CRC: 0xed9a21d2, 7188608
09-21-2023 12:29:07.758: Attempting to connect to measurement device type 49...
Segmentation fault (core dumped)

How can I connect to a Yokogawa WT310e on RHEL 9.2 Linux? What Linux OS should I use?

Thanks,
Tien
20
SERT / Re: SERT 2.0.6,2.0.7
« Last post by pgalizia on September 13, 2023, 09:08:46 AM »
Thank you, Mr.Brown, for the logs.  Those are quite helpful!

From what I see, it looks like there are two issues:
  • The LU worklet's transactions are at 90.2% of the target
  • The power sensor has 100% of its samples with unknown certainty

For the LU worklet, I see "cpuWarmupIntervalCount" in "config-all.xml" is set to "8" (instead of the default of 3).  You may need to experiment with values between 3 and 10 to find the right number of warmups for the LU worklet (I've seen cases where setting cpuWarmupIntervalCount too high resulted in a similar issue for LU, and once I reduced it one or two iterations, it was fine).

As for the power sensor running with unknown certainty, this could be caused running the sensor with auto-ranging on. which isn't supported for SERT runs (doesn't allow for uncertainty measurements).  See Section 5 in https://spec.org/sert2/SERT-userguide.pdf, which describes how to set up SERT's power ranges.  You can also check the "Power and Temperature Measurement Setup Guide" at https://spec.org/power/docs/SPEC-Power_Measurement_Setup_Guide.pdf for additional information that may be relevant to your Hioki PW3337 device.

Give those items a try and let me know if that resolves those problems.  Thanks again!
Pages: 1 [2] 3 4 ... 10