BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20121115T175000Z
DTEND:20121115T181000Z
LOCATION:155-F2
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: In the recent years, data intensive programming using Hadoop and MapReduce has become increasing important.  As normally deployed, Hadoops implementation of MapReduce in a multi-rack cluster is dependent upon the top of the rack switches and of the aggregator switches connecting multiple racks.=0A=0AWe are currently developing Hadoop-OpenFlow Enabled or Hadoop-OFE.   The basic idea is to combine OpenFlow (OF) enabled switches and a modified Hadoop JobTracker that is OpenFlow aware in order to improve the performance of Hadoop.  Over time our hope is that Hadoop-OFE over standard Ethernet can provide superior performance to Hadoop over specialized interconnects, such as InfiniBand.=0A=0AWe quantify the performance improvement of a Hadoop application called Matsu that is used for processing satellite image data when using Hadoop-OFE versus standard Hadoop.   In our Open Cloud Consortium (OCC) SC 12 Research Booth, we will also quantify the performance of Hadoop-OFE running on a wide area network between an OCC Cluster at the University of Chicago and an OCC Cluster at the AMPATH facility in Miami.
SUMMARY:OpenFlow Enabled Hadoop over Local and Wide Area Clusters
PRIORITY:3
END:VEVENT
END:VCALENDAR
BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:1.0
BEGIN:VEVENT
DTSTART:20121115T175000Z
DTEND:20121115T181000Z
LOCATION:155-F2
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: In the recent years, data intensive programming using Hadoop and MapReduce has become increasing important.  As normally deployed, Hadoops implementation of MapReduce in a multi-rack cluster is dependent upon the top of the rack switches and of the aggregator switches connecting multiple racks.=0A=0AWe are currently developing Hadoop-OpenFlow Enabled or Hadoop-OFE.   The basic idea is to combine OpenFlow (OF) enabled switches and a modified Hadoop JobTracker that is OpenFlow aware in order to improve the performance of Hadoop.  Over time our hope is that Hadoop-OFE over standard Ethernet can provide superior performance to Hadoop over specialized interconnects, such as InfiniBand.=0A=0AWe quantify the performance improvement of a Hadoop application called Matsu that is used for processing satellite image data when using Hadoop-OFE versus standard Hadoop.   In our Open Cloud Consortium (OCC) SC 12 Research Booth, we will also quantify the performance of Hadoop-OFE running on a wide area network between an OCC Cluster at the University of Chicago and an OCC Cluster at the AMPATH facility in Miami.
SUMMARY:OpenFlow Enabled Hadoop over Local and Wide Area Clusters
PRIORITY:3
END:VEVENT
END:VCALENDAR