Ann L. Drapeau

Striped tape arrays

How data striping ideas apply to arrays of magnetic tape drives is being investigated. Data striping increases throughput and reduces response time for large accesses to a storage system. Striped magnetic tape systems are particularly appealing because many inexpensive magnetic tape drives have low bandwidth. Striping may offer dramatic performance improvements for these systems. Several important issues in designing striped tape systems are considered: the choice of tape drives and robots, whether to stripe within or between robots, and the choice of the best scheme for distributing data on cartridges. One of the most troublesome problems in striped-tape arrays is the synchronization of transfers across tape drives. Another issue is how improved devices will affect the desirability of striping in the future. The results of simulations comparing the performance of striped-tape systems to nonstriped systems are presented.<<ETX>>

Striping in large tape libraries

The authors evaluate the effectiveness of applying data striping concepts to large tape libraries, for randomly distributed access to the tape library. They believe such operations will be characteristic of future tertiary storage databases using large objects, such as online libraries and multimedia databases. Using an event-driven simulator, it is shown that striped large tape libraries perform poorly for this random workload because striping causes contention for the small number of readers and robot arms in these libraries. Increasing the number of readers results in better striped performance. The authors also examine how the effectiveness of striping may change as readers and robots improve in performance. They find that striping continues to be an effective technique for increasing the throughput of large accesses if reader and robot performance scale at similar rates.

Performance and design evaluation of the RAID-II storage server

RAID-II is a high-bandwidth, network-attached storage server designed and implemented at the University of California at Berkeley. In this paper, we measure the performance of RAID-II and evaluate various architectural decisions made during the design process. We first measure the end-to-end performance of the system to be approximately 20 MB/s for both disk array reads and writes. We then perform a bottleneck analysis by examining the performance of each individual subsystem and conclude that the disk subsystem limits performance. By adding a custom interconnect board with a high-speed memory and bus system and parity engine, we are able to achieve a performance speedup of 8 to 15 over a comparative system using only off-the-shelf hardware.

Storage alternatives for video service

In the next decade, video-on-demand (VOD) services will be widely available to customers and potentially highly profitable to service providers. In order to provide access to many movies, video storage servers may contain not only magnetic disks but also high-capacity tertiary storage devices. In this paper, we study two storage device alternatives: an array of magnetic disks and an array of magnetic tapes. Magnetic disk arrays provide high-bandwidth, low-latency retrieval and moderate storage capacity at high cost. Magnetic tape arrays provide low-bandwidth, high-latency retrieval and high storage capacity at low cost. We evaluate these two storage system alternatives to determine the number of users that each can support. Our simulations show that magnetic disk arrays can support considerably more users, at a lower cost per user, than magnetic tape arrays.<<ETX>>