BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20111116T220000Z
DTEND:20111116T223000Z
LOCATION:TCC 305
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Virtualization and cloud computing make it possible to create virtual high performance computing clusters. This effort, however, is complicated by inherent differences in how these systems allocate resources.  Cloud systems immediately allocate resources or deny requests.  In contrast, parallel computing systems route all requests through a queue.  This divergence of polices hinders efforts to implement efficient, responsive, and reliable virtual clusters.=0A=0AIn this paper, we present four scheduling polices and analytical resource prediction models to estimate the resources needed to operate an efficient, responsive, and reliable virtual cluster.  We show that it is possible to estimate the size of the virtual cluster needed to provide a predictable grade of service for a high performance computing workload and estimate the queue wait time for a partial or full resource allocation.  Moreover, we show that it is possible to provide a reliable virtual cluster system using a limited pool of spare resources.
SUMMARY:Flexible Resource Allocation for Reliable Virtual Cluster Computing Systems
PRIORITY:3
END:VEVENT
END:VCALENDAR