BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20151117T231500Z
DTEND:20151118T010000Z
LOCATION:Level 4 - Lobby
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: Large-scale storage systems often use object-level replication to protect=0Aagainst server failures. This calls for the development of efficient=0Aalgorithms to map replicas to servers at scale. We introduce a modular,=0Aopen-source consistent hashing library, called libch-placement, and use it to=0Ainvestigate design trade-offs in CPU efficiency, replica declustering, and=0Aobject sharding for replicated object placement. We identify a multiring=0Ahashing configuration that strikes a balance between all three properties. We=0Aalso demonstrate that large-scale object placement can be calculated at=0Aan aggregate rate of tens of billions of objects per second with 8,192=0Aservers. These technologies can be leveraged by large-scale storage systems=0Ato improve performance, improve reliability, and manage data placement with=0Aper-object granularity.
SUMMARY:Consistent Hashing Distance Metrics for Large-Scale Object Storage
PRIORITY:3
END:VEVENT
END:VCALENDAR