BEGIN:VCALENDAR
PRODID:-//Microsoft Corporation//Outlook MIMEDIR//EN
VERSION:2.0
BEGIN:VEVENT
DTSTART:20141118T200000Z
DTEND:20141118T203000Z
LOCATION:291
DESCRIPTION;ENCODING=QUOTED-PRINTABLE:ABSTRACT: As the need for scale-out data storage in HPC continues to grow at an exponential rate, traditional data protection and availability models are being stretched past their breaking points. This talk will contrast the principal difficulties associated with hardware RAID and legacy availability models with what is made possible by instead using erasure codes to deliver a revolutionary per-file, triple parity protection storage architecture.=0A=0AMost importantly, this new approach allows data reliability to increase with the scale of the storage deployed, as opposed to decreasing as is normally expected. It also significantly improves data availability by changing from an all-up/all-down model to still maintaining partial availability in the rare case of storage systems experiencing multiple, simultaneous, drive failures.
SUMMARY:Using Erasure Codes to Revolutionize Data Reliability at Scale
PRIORITY:3
END:VEVENT
END:VCALENDAR