Thursday, October 30, 2008
12:00 pm - 1:00 pm
PLACE: CIC 2101
Microsoft Research, Cambridge, UK
Everest: scaling down I/O peaks through write off-loading
Bursts in data center workloads are a real problem for storage
subsystems. Data volumes can experience peak I/O request rates that
are over an order of magnitude higher than average load. This requires
significant over-provisioning, and often still results in significant
I/O request latency during peaks.
In this talk I'll present Everest, an architecture that allows data
written to an overloaded volume to be temporarily off-loaded into a
short-term virtual store. Everest creates the short-term store by
opportunistically pooling underutilized storage resources either on a
server or across servers within the data center. Writes are
temporarily off-loaded from overloaded volumes to lightly loaded
volumes, thereby reducing the I/O rate on the former. We evaluated
Everest using traces from a production Exchange mail server as well as
other benchmarks: our results show a 1.4--70 times reduction in mean
response times during peaks.
This is joint work with my colleagues at MSR Cambridge: Austin
Donnelly, Sameh Elnikety, Eno Thereska, and Ant Rowstron. An early
prototype of write off-loading was used to save energy by increasing
disk idle time, and was rpesented at FAST 2008. The Everest work will
be be presented at OSDI 2008.
Dushyanth Narayanan (http://www.research.microsoft.com/~dnarayan) is a
researcher in the Systems and Networking group at Microsoft Research,
Cambridge, UK. He obtained his Ph.D. from Carnegie Mellon University
in 2002. He has worked on mobile computing, performance modelling,
scalable infrastructures, power savings, and most recently on
enterprise and data center storage.
Karen Lindenfelser (8-6716)
or visit http://www.pdl.cmu.edu/SDI/