Breadcrumb
Scone, the School's HPC system
Scone consists of 30 64-bit machines, all of which run Linux 2.6.28. Access to the machines is via ssh. You must first log on the the gateway machine scone.maths.bris.ac.uk before accessing other machines. Remember to enable X-forwarding on your ssh client whilst doing this if you want to use the graphical user interface (GUI) provided by such programs as matlab.
There are eight machines for general use on the scone system. There are twelve machines for use by the statistics group all of which run under a condor job submission scheme. There are also nine machines for the applied group which do not use job submission software and of which seven are for the use of the fluids group only. Finally there is one for the use of the pure group which again does not use job submission software. A summary of the situation is shown below.
| Machine | Group | Condor | CPU | RAM |
| node1 | General | No | 8 X 2.6 GHz opteron | 32Gb Ram |
| node2 | General | No | 8 X 2.6 GHz opteron | 32Gb Ram |
| node3 | General | No | 8 X 2.6 GHz opteron | 32Gb Ram |
| node4 | General | No | 8 X 2.6 GHz opteron | 32Gb Ram |
| node5 | General | No | 8 X 2.6 GHz opteron | 64Gb Ram | node6 | General | No | 12 X 2.6 GHz Xeon | 48Gb Ram | node7 | General | No | 12 X 2.6 GHz Xeon | 48Gb Ram | node8 | General | No | 12 X 2.6 GHz Xeon | 48Gb Ram |
| zeppo | Statistics | Yes | 4 X 2.2 GHz opteron | 8Gb Ram |
| chico | Statistics | Yes | 4 X 2.2 GHz opteron | 8Gb Ram |
| harpo | Statistics | Yes | 4 X 2.2 GHz opteron | 8Gb Ram |
| groucho | Statistics | Yes | 4 X 2.2 GHz opteron | 8Gb Ram |
| barker | Statistics | Yes | 4 X 2.2 GHz opteron | 8Gb Ram |
| morecambe | Statistics | Yes | 2 X 2.6 GHz opteron | 8Gb Ram |
| wise | Statistics | Yes | 2 X 2.6 GHz opteron | 8Gb Ram |
| jake | Statistics | Yes | 8 X 2.3 GHz opteron | 16Gb Ram |
| elwood | Statistics | Yes | 8 X 2.3 GHz opteron | 16Gb Ram |
| suilven | Statistics | Yes | 12 X 3 GHz Xeon | 48Gb Ram |
| quinag | Statistics | Yes | 12 X 3 GHz Xeon | 48Gb Ram |
| canisp | Statistics | Yes | 12 X 3 GHz Xeon | 48Gb Ram |
| kelvin | Fluids | No | 4 X 2.6 GHz opteron | 8Gb Ram |
| reynolds | Fluids | No | 4 X 2.6 GHz opteron | 8Gb Ram |
| riemann | Applied | No | 4 X 2.6 GHz opteron | 16Gb Ram |
| darcy | Fluids | No | 4 X 3 GHz opteron | 12Gb Ram |
| rayleigh | Fluids | No | 4 X 3 GHz opteron | 12Gb Ram |
| hardy | Applied | No | 4 X 2.6 GHz opteron | 16Gb Ram |
| bernoulli | Fluids | No | 4 X 3 GHz opteron | 16Gb Ram |
| taylor | Fluids | No | 8 X 2.6 GHz opteron | 32Gb Ram |
| stokes | Fluids | No | 12 X 2.6 GHz Xeon | 48Gb Ram |
| heilbronn | Pure | No | 4 X 2.6 GHz opteron | 10Gb Ram |
A summary of the software installed on the scone system can be found in the Scone User Manual. Note:
- Do not run any jobs on the gateway machine. You may compile and run very small test programs. Anything else will be killed without notice.
- Any user may submit a job through condor. However members of the statistics group have a vastly increased priority.
- Only members of the appropriate group may log on to their machines directly.
Linux Farm etiquette.
- Try not to run a job that on a machine that has a load average of around 2.0 or more; definitely do not run a job on a machine with a load average of 4.0 or more. To find the load average type "w" at the command prompt. You may also use the command nodes_idle to display a summary of the idle time on the machines for general use.
- Do not run a job on a inappropriate machine (non-contributor running a job on a contributors machine)
- Do compress large files. The only way that a user can bring the whole system down is by filling up the shared space. There is no quota so that people can produce reasonable amounts of data - do not abuse this situation.
Feel free to let support know if you feel that these rules are being broken or if you have any other "turf" issues when it comes to jobs on the farm.
Using the Condor Queue
Remember if you are not in the statistics group you currently have very low priority on the condor queue.
When submitting jobs via condor it is not necessary to actually login to any of the compute servers. There are two stages to submitting a job:
- Write a file that describes the job to be submitted.
NB There is a restriction on condor jobs.
Chico, harpo, groucho and barker are "short job" machines (with a two week computation limit), whereas morecambe, wise and zeppo are "long job" machines, without this hard limit.
If you think your job will last for more than two weeks, you must specify to run it on a long job machine in your condor job submission file using:
Requirements = \
Machine =="morecambe.private2.maths.bris.ac.uk" \
|| Machine == "wise.private2.maths.bris.ac.uk" \
|| Machine == "zeppo.private2.maths.bris.ac.uk"
IF YOU DO NOT SPECIFY THIS, AND YOUR JOB LASTS FOR MORE THAN TWO WEEKS, IT WILL BE KILLED.
Here are some examples of condor files. The first runs a command from the current directory.
####################
##
## Test Condor command file
##
####################
executable = ustone6
Universe = vanilla
error = ustone6.err
output = ustone6.out
log = ustone6.log
Queue
This file tells condor that the executable "ustone6" is to be run. Standard output from this executable is to go into the file "ustone6.out' and standard error is to go into the file "ustone6.err" . The file "ustone6.log" contains any messages from the condor system (job status any error and so on). More examples can be found on the condor documentation.
