There's currently not a description on how to do this. The HPC run here at NCSU uses LSF and is static. It's static mainly because our researchers consume everything and the jobs are usually quite long > 2weeks on average.
We do (where) feasible migrate VCL blades to HPC mode during breaks, but this is usually for the summer. This is done somewhat by using the 'hpc' project listed in the VCL database, by starting a reload of the blade with a HPC compute node image. Which triggers a reload by xCAT and sets a predifined VLAN in one of the blade center's Ethernet Switch module to use the HPC vlan as the primary vlan id. There are some folks are using VCL to build short term clusters for such things. This is done using the clustering in the manage images edit image profiles tool. I think UTEP is doing this today in their VCL instance/cluster. Several researchers here at NCSU also use the image cluster feature to do this, mainly for running hadoop. Also Andy K. might be able to elaborate more, but I think with his work on the Configuration management piece could better orchestrate adding and removing nodes from a compute pool. Aaron On Fri, May 17, 2013 at 3:32 PM, Dmitri Chebotarov <[email protected]> wrote: > Hi > Is there a manual which describes how make HPC reservations with VCL ? > I see that code/database mentions 'hpc' and 'vclhpc' projects and would like > to get more info on the subj if possible. > > -- > Thank you, > > Dmitri Chebotarov > VCL Sys Eng, Engineering & Architectural Support, TSD - Ent Servers & > Messaging > 223 Aquia Building, Ffx, MSN: 1B5 > Phone: (703) 993-6175 | Fax: (703) 993-3404 > > > -- Aaron Peeler Program Manager Virtual Computing Lab NC State University All electronic mail messages in connection with State business which are sent to or received by this account are subject to the NC Public Records Law and may be disclosed to third parties.
