MPEG: a video compression standard for multimedia applications
Communications of the ACM - Special issue on digital multimedia systems
IEEE Spectrum
News on-demand for multimedia networks
MULTIMEDIA '93 Proceedings of the first ACM international conference on Multimedia
Tapes hold data, too: challenges of tuples on tertiary store
SIGMOD '93 Proceedings of the 1993 ACM SIGMOD international conference on Management of data
On multimedia repositories, personal computers, and hierarchical storage systems
MULTIMEDIA '94 Proceedings of the second ACM international conference on Multimedia
Disk-tape joins: synchronizing disk and tape access
Proceedings of the 1995 ACM SIGMETRICS joint international conference on Measurement and modeling of computer systems
Probabilistic Assignment of Movies to Storage Devices in a Video-On-Demand System
NOSSDAV '93 Proceedings of the 4th International Workshop on Network and Operating System Support for Digital Audio and Video
Using tertiary storage in video-on-demand servers
COMPCON '95 Proceedings of the 40th IEEE Computer Society International Conference
Statistical analysis of simulation output data
ANSS '76 Proceedings of the 4th symposium on Simulation of computer systems
Demand-based Document Dissemination for the World-Wide Web
Demand-based Document Dissemination for the World-Wide Web
ADC '01 Proceedings of the 12th Australasian database conference
| Hi-index | 0.00 |
Multimedia applications that are required to manipulate large collections of objects are becoming increasingly common. Moreover, the size of multimedia objects,which are already huge, are getting even bigger as the resolution of output devices improve. As a result, many multimedia storage systems are not likely to be able to keep allof their objects disk-resident. Instead a majority of the lesspopular objects have to be off-loaded to tertiary storage tokeep costs down. The speed at which objects can be accessedfrom tertiary storage is thus an important consideration. Inthis paper, we propose an adaptive data retrieval algorithmthat employs a combination of staging and direct access inservicing tertiary storage retrieval requests. At retrieval time,an object that resides in tertiary storage can either be stagedto and then played back from disks, or the object can beaccessed directly from the tertiary drives. We show that asimplistic policy that adheres strictly to staging or direct access does not exploit the full retrieval capacity of both thetertiary library and the secondary storage. To overcome theproblem, we propose a data retrieval algorithm that dynamically chooses between staging and direct access, based onthe relative load on the tertiary versus secondary devices. Aseries of simulation experiments confirms that the algorithmachieves good access times over a wide range of workloadsand resource configurations. Moreover, the algorithm is veryresponsive to changing load conditions.