Alert.png The wiki is deprecated and due to be decommissioned by the end of September 2022.
The content is being migrated to other supports, new updates will be ignored and lost.
If needed you can get in touch with EGI SDIS team using operations @ egi.eu.

FAQ HEP SPEC06

From EGIWiki
(Redirected from FAQ01)
Jump to navigation Jump to search
Main EGI.eu operations services Support Documentation Tools Activities Performance Technology Catch-all Services Resource Allocation Security


Documentation menu: Home Manuals Procedures Training Other Contact For: VO managers Administrators



Transition to HEP SPEC, a new CPU benchmark

Q1: Why adopting HEP SPEC 06?

The traditional si2k CPU benchmark is now obsolete and it is time to move to HEP SPEC, a new CPU benchmark that will replace si2k and will become the reference benchmark for accounting purposes.

Have a look at the "Transition to a new CPU benchmarking unit for the WLCG" for a detailed description of “why".

Q2: What is HEP SPEC 06?

The HEP-SPEC06 benchmark is designed to scale with the performances of the high-energy physics codes on similar machines. The goal was to have an accuracy of ± 5% but for the moment the agreement is significantly higher.

The measurement of HEP-SPEC repeated on identical machines varies less than 1%. If the computing machines are similar, i.e. same processors and at least 2 GB per core, the results obtained are very close, within some percent, so that it is unnecessary to perform measures on all computing hosts. It is enough to do the measurement for one type of processor and consider it valid for all the machines with the same processor.

If you are using different OS and specially different compilers, the data will change.

Q3: Where can I find information about HEP SPEC 06 measurements?

Some example results are available on the HEPIX group-page at CASPUR: HEPIX Bench-Results, where one can see the differences between gcc3.4.x and gcc4.1.x. Other results - available on the wiki-page HEPIX - CASPUR

Additional results tables are available from various EGI partners:

If you don't find your computing machine in that table, then it is better to try to do the measurement because extrapolating the results increases further the error.

Q4: How can I run the HEP SPEC 06 benchmark?

If you want to make HEP-SPEC06 on your own own, detailed instructions are available at: CERN wiki In Short you need the following:

  1. A machine with any version of Linux compatible with Scientific Linux (RHEL, SL, SLC, CentOS)
  2. The gcc compiler should be installed
  3. Configuration files and run script (available as a gzipped tar archive from the CERN Wiki or from here). The archive's md5sum is 9fed92b8d515b88904705f76809c4028
  4. A tar ball of the SPECcpu2006 DVD called SPEC2006_v11.tar.bz2 that should be in the same directory as the run script

Q5: My site already adopted HEP SPEC 06. Do I still need to publish SpecInt2000?

The transition to HEP-SPEC does not eliminate the need to publish the computing power in SpecInt2000 (due to backward compatibility with sites not publishing yet HEP-SPEC). In this case you may calculate the value SpecInt2000 starting from HEP-SPEC through the following relation:

  • value_kSI2K = value_HEP-SPEC / 4 (or value_HEP-SPEC = 4 * value_kSI2K)

For the GlueHostBenchmarkSI00 attribute in the GLUE v1.3 schema the following relation is easier to use:

  • value_SI00 = value_HEP-SPEC * 250 rounded to the nearest integer

Q6: How are HEP SPEC 06 results set in YAIM?

The YAIM variable CE_OTHERDESCR is used to set the GlueHostProcessorOtherDescription attribute. The value of this variable MUST be defined in your site-info.def file as:

Cores=<CE_LOGCPU/CE_PHYSICALCPU> [, Benchmark=<value>-HEP-SPEC06]

where the ratio CE_LOGCPU / CE_PHYSICALCPU means the average number of cores per physical CPU in a SubCluster; in the case of (slightly) heterogeneous subclusters it could be non-integer. The second value of this attribute MUST be published only in the case the CPU power of the SubCluster has been computed using the HEP-SPEC06 benchmark. The Benchmark value must be the average HEP_SPEC06 result per core, in the SubCluster.

These variables are set in your site-info.def file. After this, the variables need to be published by the CE's resource BDII, configured e.g. by standard YAIM commands.

The total CPU capacity of the cluster is computed as Benchmark * CE_LOGCPU.


Tferrari 10:38, 14 February 2011 (UTC)