Estimating job start times on workload management systems

Abstract

A method for estimating job start times on a system may include: performing a first simulation of the system by a first process to generate a first estimate of the start time based a first snapshot of the computing system; and performing a second simulation of the system to generate a second estimate of the start time. The first and second estimates may each be either a distinct time estimate or a range of time estimate, depending on whether the simulations dispatches the job before the respective first and second processes end. The method may then include generating a fifth estimate of the time that the computing job will start by a third process based on estimates respectively generated by the first and second processes.

What is claimed is: 
     
         1 . A method for estimating a time that a computing job will start executing on a system having a cluster of computers, comprising:
 performing a first simulation of the system by a first process, the first process executing for a first execution time to generate either a first or a second estimate of the time that the computing job will start based on a first snapshot of the system, wherein the first estimate specifies a distinct time and the second estimate specifies a range of times, the first process generating:
 the first estimate when the first simulation dispatches the computing job before the first simulation ends, and 
 the second estimate when the first simulation fails to dispatch the computing job before the first simulation ends; 
   performing a second simulation of the system by a second process, the second process executing for a second execution time to generate either a third or a fourth estimate of the time that the computing job will start based on a second snapshot of the system, wherein the third estimate specifies a distinct time and the fourth estimate specifies a range of times, the second process generating:
 the third estimate when the second simulation dispatches the computing job the second simulation ends, and 
 the fourth estimate when second simulation fails to dispatch the computing job before the second simulation ends, wherein first and second processes execute in parallel and the second execution time is longer than the first execution time; 
   prioritizing the estimates generated by the first and second processes, wherein:
 distinct time estimates are prioritized higher than range of times estimates, 
 estimates generated at first time are prioritized lower than estimates generated at a second time when the second time occurs subsequent to the first time, and 
 estimates generated from simulations based on snapshots of the system obtained at a first time are prioritized lower than estimates generated by simulations based on snapshots of the system obtained at a second time when the second time occurs subsequent to the first time; and 
   generating a fifth estimate of the time that the computing job will start by a third process based on the priorities of the estimates respectively generated by the first and second processes.