hpc Archives - Page 7 of 25

MPI affinity

Aug 6, 2015

MPI socket vs core affinity When running a MPI job you can bind your threads to either a socket or to individual cores. There is no one best solution, the choice depends completely on your task and the way it…

Berkeley Lab Checkpoint/Restart

Jul 7, 2015

hpc, SLURM

Berkeley Lab Checkpoint/Restart has been installed on the SLURM cluster. This allows users to checkpoint a job, cancel it and then resume the job at a later date. The executable is started with the cr_run wrapper: cr_run /home/andy/ram.pl >> /home/andy/ramtest.out…

SLURM and memory management

Jun 26, 2015

hpc, SLURM

SLURM allows a DefMemPerCPU and a MaxMemPerCPU to be set. If a user does not set a memory limit the default will be used. This is normally set to MaxMem/NumCores. As memory is a consumable resource (SelectTypeParameters=CR_Core_Memory) MaxMemPerCPU serves not

…

GPUs and GRES

Jun 22, 2015

GPU, hardware, hpc, SLURM

Our current cluster, hex, runs Torque with MAUI as the scheduler. While MAUI is GPU aware it does not allow GPUs to be scheduled. In other words you can list the nodes with GPUs but you cannot submit a job …

SLURM job preemption

Jun 19, 2015

hpc, SLURM

SLURM provides a preemption mechanism to deal with situations where cluster become overloaded. This can be configured in several ways:

FIFO:
This is the most simplistic method of queueing in which there is no preemption, jobs come in, queue and …

Where are all the HPC servers disappearing to?

Jun 3, 2015

hpc, SLURM

Those users still making use of hpc.uct.ac.za will have noticed that a few worker nodes have vanished. As mentioned previously we’re investigating a new scheduler, SLURM, Simple Linux Utility for Resource Management. SLURM is a very different animal to

…

Hardware Donation to the University of the Western Cape

May 11, 2015

hpc

During 2014 we decommissioned the “HP BL20p gen3/gen4” series of blades we used for our first HPC cluster. The hardware is still in good condition and still has a good few years of computing it could deliver. We decided to …

New stuff

Apr 30, 2015

hardware, hpc, SLURM

SLURM

The HPC team is busy testing a prototype SLURM cluster. This will replace both the hex and hpc clusters, although it is likely that the hex infrastructure will be incorporated into the new cluster. The time frame for

…

HPC training

Feb 27, 2015

hpc

The eResearch team will be running several HPC training courses this year, basic linux skills for HPC and advance HPC concepts. The first round of courses for February and March were booked up before we could advertise them here, however …

2014 Summary

Dec 29, 2014

hpc

This year has been a very productive one for UCT HPC. We completed more CPU hours than all previous years combined and expanded our cluster by more than doubling our core count and also adding a new GPU server with …

hpc

UCTHPC