Intel Research Seminar

DATE: Thursday, March 6, 2003
TIME: Noon - 1:30 pm
PLACE: Intel Seminar (417 S. Craig Street - 3rd Floor)

Eran Gabber
Bell Labs

StarFish: Highly-available Block Storage

The world's growing dependency on data increases the need for highly-available data storage. Geographically-dispersed replicas can provide data availability despite a major site Failure, and are increasingly affordable given the price-performance trends of storage devices, servers, and high-speed networks. In this light we present StarFish, a block storage system built from commodity servers running FreeBSD, which are connected by standard high-speed IP networking gear. StarFish achieves high availability by transparently replicating data over multiple storage sites. StarFish is accessed via a host-site appliance that masquerades as a host-attached storage device, hence it requires no special hardware or software in the host computer.

We show that a StarFish system with 3 replicas and a write quorum size of 2 is a good choice, based on a formal analysis of data availability and reliability: 3 replicas with individual availability of 99% and a write quorum of 2 gives better than 99.9999% overall data availability.

Although StarFish increases the per-request latency relative to a direct-attached RAID, we show how to design a highly-available StarFish configuration that provides most of the performance of a direct-attached RAID on an I/O-intensive benchmark, even during the recovery of a failed replica. Moreover, the third replica may be connected by a link with long delays and limited bandwidth, which alleviates the necessity of dedicated communication links to all replicas.

Eran Gabber received his Ph.D. in Computer Science from Tel-Aviv University, Israel, in 1993. From 1993 to 1995 he was the principal researcher of nSOF Parallel Software, Ltd., a startup company in Israel. He joined Bell Laboratories (first a part of AT&T, later a part of Lucent Technologies) in 1995, where he is currently a distinguished member of technical staff. His research interests include operating systems, file and storage systems, tools for building specialized systems, quality of service, network management, Internet privacy, and electronic commerce.

For Further Seminar Info:
Contact Kim Kaan, 412-605-1203, or visit

SDI Home: