HECToR Monthly Report, November 2008
Information on the utilisation, disk allocations, slowdowns and helpdesk statistics can be found in the associated SAFE monthly report.
Dates covered: 08:00 1 November 2008 to 08:00 1 December 2008
Number of hours: 720
1: Availability
Scheduled down time: 11 hours 41 minutes.
Incidents
The following incidents were recorded:
| Severity | Number |
| 1 | 3 |
| 2 | 0 |
| 3 | 22 |
| 4 | 0 |
Of the four severity levels, level 1 corresponds to a contractual failure.
All of the above SEV-3 Incidents were attributed to single node failures.
Details of severity level 1 incidents
| ID | Date | Description | Length | Attribution |
| Incident-464 | 12/11/2008 | System failure after lustre errors | 02:33 | Cray |
| Incident-477 | 25/11/2008 | Link inactive problem | 01:46 | Cray |
| Incident-480 | 27/11/2008 | Voltage fault on mezz card | 01:25 | Cray |
MTBF and Serviceability
| Attribution | Failures | MTBF | UDT | Serviceability |
| Cray | 3 | 244 | 05:44:00 | 99.2% |
| Site | 0 | ∞ | 00:00:00 | 100% |
| External | 0 | ∞ | 00:00:00 | 100% |
| Other | 0 | ∞ | 00:00:00 | 100% |
| Overall | 3 | 244 | 05:44:00 | 99.2% |
- 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 XT single node failures
| Error Type | Number |
| RX message header CRC Error | 11 |
| MCA bk0 error (Internal Opteron) | 2 |
| UME bk0/bk4 (Dimms) | 6 |
| Seastar trap 1 error (PPC data cache parity error) | 1 |
| Node did not come back after reboot | 1 |
Details of X2 single node failures
| Error Type | Number |
| Software bug | 1 (triple nodes) |
2: Courses
This information is supplied by NAG Ltd
| Title of Course | Dates | Available places | Ordinary attendees | Paying attendees | CSE Staff | Total attending |
| Introduction to MPI, NAG Manchester | 11-12 November 2008 | 16 | 11 | 3 | 1 | 15 |
| Advanced MPI, NAG Manchester | 13 November 2008 | 16 | 7 | 2 | 1 | 10 |
| MPI One-Sided Communication and MPI-IO, NAG Manchester (1) | 14 November 2008 | 16 | 0 | 0 | 0 | 0 |
| OpenMP and Mixed-Mode Programming, NAG Manchester | 25 - 26 November, 2008 | 16 | 6 | 1 | 1 | 8 |
(1) Cancelled. Two people registered but later withdrew.
3: Quality tokens
| 16 November 2008 21:39:59 | Mr Andrea Spitaleri | **** | I am very happy so far with the hector service. |
| 26 November 2008 14:05:31 | Mr Hristo Iliev | **** | How about some 24hrs parallel queues? |
4: Hours worked
| Group | Days worked | FTEs |
| USL | 65.1 | 3.7 |
| OSG | 76.3 | 4.3 |
5: Performance metrics
Technology Provision
| Description | TSL | FSL | Value |
| Technology reliability | 85% | 98.5% | 99.2% |
| Technology throughput | 7000 hours | 8367 hours | 8575 hours |
| Capability job completion rate | 70% | 90% | 92 % |
| Technology MTBF | 100 hours | 126.4 hours | 244 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% | 100% |
| Number of SP FTEs | 7.3 | 8.0 | 8.7 | 8.0 |
| SP serviceability | 80% | 99% | 99.5% | 100% |
