BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20121115T173000Z
DTEND:20121115T180000Z
LOCATION:355-D
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Evaluating new ideas for job scheduling or data transfer algorithms=0Ain large-scale grid systems is known to be notoriously challenging.=0AExisting grid simulators expect=0Ato receive a realistic workload as an input. Such input is difficult to provide=0Ain absence of an in-depth study of representative grid workloads.=0A=0AIn this work, we analyze the ATLAS workload processed=0Aon the resources of NDG Facility. ATLAS is one of the biggest grid technology users, with extreme demands for CPU=0Apower and bandwidth. =0AThe analysis is based on the data sample with ~1.6 million jobs,=0A1723TB of data transfer, and 873 years of processor time.=0AOur additional contributions are (a) scalable workload models that can be used to=0Agenerate a synthetic workload for a given number of jobs, (b) an open-source workload=0Agenerator software integrated with existing grid simulators, and (c) suggestions for grid system designers based on the insights of data analysis.
SUMMARY:ATLAS Grid Workload on NDGF Resources: Analysis, Modeling, and Workload Generation
PRIORITY:3
END:VEVENT
END:VCALENDAR
BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20121115T173000Z
DTEND:20121115T180000Z
LOCATION:355-D
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Evaluating new ideas for job scheduling or data transfer algorithms=0Ain large-scale grid systems is known to be notoriously challenging.=0AExisting grid simulators expect=0Ato receive a realistic workload as an input. Such input is difficult to provide=0Ain absence of an in-depth study of representative grid workloads.=0A=0AIn this work, we analyze the ATLAS workload processed=0Aon the resources of NDG Facility. ATLAS is one of the biggest grid technology users, with extreme demands for CPU=0Apower and bandwidth. =0AThe analysis is based on the data sample with ~1.6 million jobs,=0A1723TB of data transfer, and 873 years of processor time.=0AOur additional contributions are (a) scalable workload models that can be used to=0Agenerate a synthetic workload for a given number of jobs, (b) an open-source workload=0Agenerator software integrated with existing grid simulators, and (c) suggestions for grid system designers based on the insights of data analysis.
SUMMARY:ATLAS Grid Workload on NDGF Resources: Analysis, Modeling, and Workload Generation
PRIORITY:3
END:VEVENT
END:VCALENDAR