Changes between Version 12 and Version 13 of JobIn
 Timestamp:
 Jul 13, 2011, 3:23:56 PM (10 years ago)
Legend:
 Unmodified
 Added
 Removed
 Modified

JobIn
v12 v13 26 26 === Resource estimates and bounds === #resources 27 27 28 It's important that you supply accurate values for these parameters. 29 See [JobEst estimating job sizes] for tips on how to do so. 30 28 31 '''rsc_fpops_est''':: 29 An estimate of the average number of floatingpoint operations required to complete the computation. This is used to estimate how long the computationwill take on a given host.32 An estimate of the number of floatingpoint operations required to complete the job, used to estimate how long the job will take on a given host. 30 33 '''rsc_fpops_bound''':: 31 A bound on the number of floatingpoint operations required to complete the computation. If this bound is exceeded, the applicationwill be aborted.34 An upper bound on the number of floatingpoint operations required to complete the job. If this bound is exceeded, the job will be aborted. 32 35 '''rsc_memory_bound''':: 33 An estimate of application's largest working set size. The workunit will only be sent to hosts with at least this much available RAM.36 An estimate of job's largest working set size. The job will only be sent to hosts with at least this much available RAM. If this bound is exceeded, the job will be aborted. 34 37 '''rsc_disk_bound''':: 35 A bound on the maximum disk space used by the application, including all input, temporary, and output files. The workunit will only be sent to hosts with at least this much available disk space. If this bound is exceeded, the applicationwill be aborted.38 A bound on the maximum disk space used by the job, including all input, temporary, and output files. The job will only be sent to hosts with at least this much available disk space. If this bound is exceeded, the job will be aborted. 36 39 '''rsc_bandwidth_bound''':: 37 If nonzero, this job will be sent only to hosts with at least this much download bandwidth. Use for applications with very large input files. 38 39 It's important that you supply accurate values for these parameters. To get initial estimates, run your application on your own machines, and monitor the memory and disk usage. To view the distribution of FLOPs, use the script html/ops/job_times.php (the '''FLOP count statistics''' link on your project's [HtmlOps administrative web interface]). 40 If nonzero, this job will be sent only to hosts with at least this much download bandwidth. Use for jobs with very large input files. 40 41 41 42 === Redundancy and scheduling attributes === #scheduling