While I’m not convinced that the “number of jobs” is very useful as a metric (what if they were all aborted in less than a second because of script errors? They weren’t in this case, but I’m just saying…), the sheer scale of the number of jobs run by TACC’s Ranger is worth noticing
The Ranger supercomputer, one of the most powerful systems in the world for open science research, has run about 1.1 million jobs in under two years.
…Bill Barth, director of TACC’s HPC group, said, “The demand for time on Ranger has been very high and instrumental to making TeraGrid the nation’s largest resource for open science computational research. The system has run more than 600 million central processing unit hours so far.”
As for the user who ran the millionth job, Barth said it was a small post-processing job (16 processors) completed by Dr. Yonghui Weng, research associate, in Professor Fuqing Zhang’s hurricane research group at the Pennsylvania State University Department of Meteorology.
Ranger, built by Sun with AMD Opterons, is a number 8 on the Top500 heading into SC09.