| \section{Performance Study} |
| |
| \begin{figure}[htb] |
| \centerline{\epsfig{file=../figures/times.eps}} |
| \caption{Time to execute /bin/hostname with various node counts} |
| \label{timing} |
| \end{figure} |
| |
| We were able to perform some SLURM tests on a 1000 node cluster at LLNL. |
| Some development was still underway at that time and |
| tuning had not been performed. The results for executing simple 'hostname' program |
| on two tasks per node and various node counts is show |
| in Figure~\ref{timing}. We found SLURM performance to be comparable |
| to the Quadrics Resource Management System (RMS)~\cite{RMS} |
| for all job sizes and about 80 times faster than IBM |
| LoadLeveler~\cite{LoadLevelerWeb,LoadLevelerManual} at tested job sizes. |
