Cloud & VM status
From GridPP Wiki
Revision as of 18:53, 12 March 2015 by Andrew Mcnab 2ef6d6672c (Talk | contribs)
This page will collect information from GridPP sites regarding their current deployment of job execution systems providing VMs, including Clouds. The information will help with wider considerations and strategy.
For sites with multiple VM Provider / VM Lifecycle Manager implementations, please create multiple rows.
Site | VM Provider | VM Lifecycle Manager | Experiment/VO | APEL accounting | Notes | |||||
GridPP DIRAC | ALICE | ATLAS | CMS | LHCb | Others (specify) | |||||
RAL Tier-1 | HTCondor | Condor Vacuum | X | X | X | X |
|
|||
RAL Tier-1 | OpenNebula | |||||||||
Brunel | OpenVZ |
|
||||||||
Imperial | OpenStack | CloudScheduler | X |
| ||||||
QMUL | Gridengine (local) | None | Son of Gridengine | CREAM | No | Yes | Deploy cloudstack, find scalable solution to get our storage usable in the cloud | |||
RHUL | Torque/Maui (local) | Torque/Maui support non-existent | Will follow the consensus | CREAM | No | Yes |
| |||
UCL | No | X |
| |||||||
Lancaster | Son of Gridengine (HEC), torque/maui (local) | Disillusioned with torque/maui. | Slurm or HTCondor. | CREAM, interested in ARC | No | Yes | VMWare testing; Vac in production | |||
Liverpool (Single core cluster) | Torque Maui (local) | Poor Support, Maui intrinsically broken | Cream | No | NO | None | ||||
Manchester | Torque/Maui (local) | Maui is unsupported. It had memory leaks. Robert wrote a patch and there was nowhere to feed it back into. | HTCondor | Currently CREAM, investigating ARC-CE | No | Yes | Vac in production |
| ||
Sheffield | Torque/Maui (local) | Torque/Maui support non-existent | HTCondor is in testing mode | CREAM CE, ACR CE is in test | No | Yes |
| |||
Durham | SLURM (local) | No reason | ARC CE | No | Yes | N/A |
| |||
Edinburgh | Gridengine | None | No reason | Cream CE for standard production, ARC CE for exploratory HPC work | No | Yes |
| |||
Glasgow | HTcondor (local), Torque/Maui (local) | Becomes unresponsive at times of high load or nodes being un-contactable. | Investigating HTCondor/SoGE/SLURM as a replacement. | ARC, Cream | Yes | Yes | N/A | |||
Birmingham | Torque/Maui | Maui sometimes fails to see new jobs and so nothing is scheduled | HTCondor | CREAM | No | No | Testing Vac setup |
| ||
Bristol | HTCondor (shared), torque + maui (local) | None | No reason | ARC & CREAM CEs | No | X |
| |||
Cambridge | Torque/Maui (local) | Torque/Maui support non-existent | Will follow the consensus | CREAM CE | No | Yes | None at present | |||
Oxford | HTCondor (local) | None | No reason | ARC CE in production | Yes | Yes | OpenStack in production. Testing VAC |
| ||
RAL PPD | HTCondor | None | No reason | ARC CE | Yes | Yes |
| |||
Sussex | (Shared) Gridengine - (Univa Grid Engine) | None | No reason | CREAMCE | Looking into it | Yes |
|