Skip to Main Content
Efficient disk storage is a crucial component formany applications. The commonly used method of storing data on disk using file systems or databases incurs significant overhead which can be a problem for applications which need to frequently access and update a large number of objects. This paper presents efficient algorithms for managing persistent storage which usually only require a single seek for allocations and deallocations and allow the state of the system to be fully recoverable in the event of a failure. Our system has been deployed for persistently storing data at the most accessed sport and event Web site hosted by IBM and results in considerable performance improvements over databases and file systems forWeb-related workloads.
Date of Conference: 17-20 April 2001