Energy Efficient Selective Cache Invalidation

In a mobile environment, users of portable computers can query databases over the wireless communication channels. To reduce contention on the limited bandwidth of the wireless channels, frequently accessed data are cached at the mobile clients. However, maintenance of the cache consistency is complicated by the mobile clients’ frequent disconnection to conserve energy. One promising approach in the literature is for the server to periodically broadcast invalidation reports from which clients can salvage their cache content that are still valid. This mechanism is, however, not energy efficient as clients are expected to examine the entire invalidation report. In this paper, we reexamine the cache coherency problem and propose three novel cache coherency schemes. While these schemes are based on periodic broadcast of invalidation reports, they allow clients to selectively tune to the portions of the invalidation report that are of interest to them. This allows the clients to minimize the power consumption when invalidating their cache content. We conducted extensive studies based on a simulation model. Our study shows that the proposed schemes are not only effective in salvaging the cache content that are still valid (and hence result in lower access time), but are also efficient in energy utilization. While none of the proposed algorithms is superior in both the access time and energy efficiency, one of the schemes, Selective Cache Invalidation, provides the best overall performance.

[1]  Robert W. Brodersen,et al.  A Portable Multimedia Terminal for Personal Communications , 1992 .

[2]  Jun Cai,et al.  Broadcast-Based Group Invalidation: An Energy-Efficient Cache Invalidation Strategy , 1997, Inf. Sci..

[3]  Jeffrey Xu Yu,et al.  Energy efficient filtering of nonuniform broadcast , 1996, Proceedings of 16th International Conference on Distributed Computing Systems.

[4]  Tomasz Imielinski,et al.  Energy efficient indexing on air , 1994, SIGMOD '94.

[5]  Rafael Alonso,et al.  Query Optimization for Energy Efficiency in Mobile Environments , 1993, FMLDO.

[6]  Beng Chin Ooi,et al.  On incremental cache coherency schemes in mobile computing environments , 1997, Proceedings 13th International Conference on Data Engineering.

[7]  Philip S. Yu,et al.  Energy-efficient caching for wireless mobile computing , 1996, Proceedings of the Twelfth International Conference on Data Engineering.

[8]  G. Komoriya,et al.  Hobbit: a high-performance, low-power microprocessor , 1993, Digest of Papers. Compcon Spring.

[9]  John Zahorjan,et al.  The challenges of mobile computing , 1994, Computer.

[10]  Dik Lun Lee,et al.  Using Signature and Caching Techniques for Information Filtering in Wireless and Mobile Environments , 1996 .

[11]  Ahmed K. Elmagarmid,et al.  Bit-Sequences: An adaptive cache invalidation method in mobile client/server environments , 1997, Mob. Networks Appl..

[12]  Kevin Watkins Discrete Event Simulation in C , 1993 .

[13]  Tomasz Imielinski,et al.  Power Efficient Filtering of Data an Air , 1994, EDBT.

[14]  Tomasz Imielinski,et al.  Mobile wireless computing: challenges in data management , 1994, CACM.

[15]  B. R. Badrinath,et al.  To send or not to send: implementing deferred transmissions in a mobile host , 1996, Proceedings of 16th International Conference on Distributed Computing Systems.

[16]  Philip S. Yu,et al.  Indexed sequential data broadcasting in wireless mobile computing , 1997, Proceedings of 17th International Conference on Distributed Computing Systems.