HECToR Monthly Report, October 2009
Information on the utilisation, disk allocations, slowdowns and helpdesk statistics can be found in the associated SAFE monthly report.
Dates covered: 08:00 1 October 2009 to 08:00 1 November 2009
Number of hours: 744
1: Availability
Scheduled down time: 15 hours 00 minutes.
Incidents
The following incidents were recorded:
Severity | Number |
1 | 0 |
2 | 1 |
3 | 14 |
4 | 0 |
Of the four severity levels, level 1 corresponds to a contractual failure.
Out of the 14 SEV-3 Incidents, 14 were attributed to single node failures.
Details of severity level 1 incidents
None this month.MTBF and Serviceability
Attribution | Failures | MTBF | UDT | Serviceability |
Cray | 0 | ∞ | 00:00:00 | 100% |
Site | 0 | ∞ | 00:00:00 | 100% |
External | 0 | ∞ | 00:00:00 | 100% |
Other | 0 | ∞ | 00:00:00 | 100% |
Overall | 0 | ∞ | 00:00:00 | 100% |
- Note 1: Serviceability%= 100*(WCT-SDT-UDT)/(WCT-SDT)
- Note 2: MTBF (Mean Time Between Failures) is defined as 732/Number of failures.
Details of single node failures
Error Type | Number |
UME bk0/bk4 (Dimms) | 5 |
UME Bank 1 (Internal Opteron) | 1 |
RX message header CRC Error | 7 |
RX packet sequence number error | 1 |
2: Courses
This information is supplied by NAG LtdTitle of Course | Dates | Available Places | Ordinary Attendees | Paying Attendees | CSE Staff | Total Attending |
Advanced MPI, NAG Manchester | 7 October 2009 | 12 | 2 | 0 | 1 | 3 |
Parallel IO, NAG Manchester | 8 October 2009 | 12 | 3 | 0 | 2 | 5 |
Best Practice in HPC Software Development, NAG Oxford | 12 - 16 October 2009 | 12 | 6 | 0 | 0 | 6 |
Best Practice in HPC Software Development, NAG Manchester | 19 -23 October 2009 | 12 | 9 | 0 | 2 | 11 |
Introduction to HECToR, University of Oxford | 22 October 2009 | 12 | 9 | 0 | 0 | 9 |
3: Quality Tokens
Date | Tokens Awarded | Comment | Consortium |
07-Oct-2009 15:29:41 | • • • | Several unexplained shutdowns, a damaged node and long maintenance sessions. People run experiments during the week but weekends the machine is generally left a lot emptier, it would make sense to do work restoring nodes on a Sunday not a Wednesday. | n02 |
07-Oct-2009 16:05:35 | • • | HECToR has recently had unpredictable and planned outages. The weekly maintenance, may be convenient for the engineers, but being in the middle of the week is exactly when scientists are working? | n02 |
15-Oct-2009 14:20:53 | * * * * * | great support | e90 |
30-Oct-2009 10:53:31 | * * * * | N/A | e24 |
4: Hours Worked
Group | Days worked | FTEs |
USL | 76.7 | 4.3 |
OSG | 81.8 | 4.6 |
5: Performance Metrics
Technology Provision
Description | TSL | FSL | Value |
Technology reliability | 85% | 98.5% | 100% |
Technology throughput | 7000 hours | 8367 hours | 8604 hours |
Capability job completion rate | 70% | 90% | 100% |
Technology MTBF | 100 hours | 126.4 hours | ∞ |
Note: Technology throughput is calculated: 12*(732-UDT-SDT); 732 - annual average number of hours in a month
Note: MTBF is calculated as 732/number of failures
Service Provision
Description | TSL | FSL | USL | Value |
Percentage of non-in-depth queries resolved within one day | 85% | 97% | 99% | 98.8% |
Number of SP FTEs | 7.3 | 8.0 | 8.7 | 8.9 |
SP Serviceability | 80% | 99% | 99.5% | 100% |