Tertiary storage: an evaluation of new applications
Tertiary storage: an evaluation of new applications
Random I/O scheduling in online tertiary storage systems
SIGMOD '96 Proceedings of the 1996 ACM SIGMOD international conference on Management of data
Hierarchical Storage Management in a Distributed VOD System
IEEE MultiMedia
A Cost-effective Near-line Storage Server for Multimedia System
ICDE '95 Proceedings of the Eleventh International Conference on Data Engineering
Scheduling Tertiary I/O in Database Applications
DEXA '97 Proceedings of the 8th International Workshop on Database and Expert Systems Applications
Hierarchical Scheduling Algorithms for Near-Line Tape Libraries
DEXA '99 Proceedings of the 10th International Workshop on Database & Expert Systems Applications
Video Server with Tertiary Storage
MSS '01 Proceedings of the Eighteenth IEEE Symposium on Mass Storage Systems and Technologies
Scheduling and Data Replication to Improve Tape Jukebox Performance
ICDE '99 Proceedings of the 15th International Conference on Data Engineering
Scheduling: Theory, Algorithms, and Systems
Scheduling: Theory, Algorithms, and Systems
Implementing and Evaluating Jukebox Schedulers Using JukeTools
MSS '03 Proceedings of the 20 th IEEE/11 th NASA Goddard Conference on Mass Storage Systems and Technologies (MSS'03)
| Hi-index | 0.00 |
We present a hierarchical multimedia archive that can serve complex multimedia requests from tertiary storage. Requests can consist of multiple request units of streamed and non-streamed data. The request units can have arbitrary synchronization patterns.Our scheduler Promote-IT promotes data from tertiary to secondary storage with real-time guarantees. Promote-IT uses an on-line heuristic algorithm to compute feasible schedules and a separate ASAP dispatcher to increase the efficiency of the resource usage. The heuristic algorithm runs in polynomial time. Schedules are optimized to give short response times to incoming requests.Three major problems complicate this scheduling problem. First, the fragments of requested real-time data and their synchronization are unpredictable. Second, the medium switching times in tertiary storage are high, and the number of drives and robots is low compared to the number of removable media. Third, the shared resources in the tertiary storage system create resource contention problems.