BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20151119T163000Z
DTEND:20151119T170000Z
LOCATION:19AB
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: The job management system is the HPC middleware responsible for distributing computing power to applications. While such systems generate an ever increasing amount of data, they are characterized by uncertainties on some parameters like the job running times. The question raised in this work is: To what extent is it possible/useful to take into account predictions on the job running times for improving the global scheduling?=0A=0AWe present a comprehensive study for answering this question assuming the popular EASY backfilling policy. More precisely, we rely on some classical methods in machine learning and propose new cost functions well-adapted to the problem. Then, we assess our proposed solutions through intensive simulations using several production logs. Finally, we propose a new scheduling algorithm that outperforms the=0Apopular EASY backfilling algorithm by 28% considering the average bounded slowdown objective.
SUMMARY:Improving Backfilling by using Machine Learning to Predict Running Times
PRIORITY:3
END:VEVENT
END:VCALENDAR