Figure 4
From: High throughput profile-profile based fold recognition for the entire human proteome
CPU usage. Number of CPUs used from each cluster at each independent Grid domain (cs.ucl.ac.uk, ccc.ucl.ac.uk and lesc.imperial.ac.uk), calculated as the number of individual jobs running at each cluster per hourly interval. For the majority of the experiment the total number of CPUs remained over 500. The tail end of the experiment is shown where the number of CPUs available begins to exceed the number of sequences left to process.