| S. Sarawagi and M. Stonebraker, "Single Query Optimization for Tertiary Memory," Technical Report UCB:S2K-94-45, Univ. of California, Berkeley, 1994. |
....designed schemes for handling this type of query execution. Instead, we are trying to optimize execution dynamically at run time. We believe that these techniques can be extended to handle both indices and relations on tape, especially if they are combined with the mechanisms described in [SS94] and [SS96] The remainder of this paper is organized as follows. In Section 2, we summarize research related to the problem of adding tertiary storage support to database systems. The mechanisms used to extend Paradise to handle tertiary storage volumes are described in Section 3. Section 4 ....
....collecting the access patterns and reorganizing data on tapes over time is a difficult task to accomplish in an online system. Our approach puts emphasis on optimizing run time query execution, in a manner that is independent of the original data placement. Tertiary Storage Query Processing [SS94] and [ML95, ML96] propose special techniques to optimize the execution of single join operations for relations stored on tape. Careful selection of the processing block size and the ordering of block accesses is demonstrated to reduce execution time by about a factor of 10. ML96] exploits the ....
S. Sarawagi and M. Stonebraker. "Single Query Optimization for Tertiary Memory," Proc. of the 1994 SIGMOD Conference, May, 1994.
....CPU time for robotically mounting a tape. t hmount : average CPU time for manually mounting a tape. taperate i : rate of transfer of data by the tape drives. For the tape drives considered here taperate i = 2:5MB=sec. t tseek : average tape seek time (equal to 1=3 of the maximum seek time [17]) blocksize d : block size used by the disks which is 16KB. blocksize t : block size used by the tapes which is 15KB. B.2 Computation of Service Demands This section gives the equations used to compute the service demands at each device of the queuing model using the parameters described in the ....
Sunita Sarawagi and Michael Stonebraker, "Single Query Optimization for Tertiary Memory", Tech. Rep. UCB:S2K-94-45, University of California at Berkeley, 1994.
....multi page I O requests, regardless of request size. A detailed cost model which combines both cost metrics has been considered in [5] A tape join study done as part of the Sequioa 2000 project [13] assumes a configuration of two tapes and one tape drive, and hence focuses on media switch delays [11]. The study employs a transfer only cost model for tape drives, counting the number of chunks, or extents, transferred. CPU and disk I O costs are ignored in the cost model. Memory size, relative to the size of relations, is typically a key factor in selecting a join method. For disk based ....
S. Sarawagi and M. Stonebraker. Single query optimization for tertiary memory. Technical Report 45, Univ. of California at Berkeley, Mar. 1994.
....database systems in which ad hoc queries can make predetermining access patterns essentially impossible. In addition, collecting the access patterns and reorganizing data on tapes over time may be a difficult task to accomplish in an on line system. Tertiary Storage Query Processing [22] and [23, 24] propose special techniques to optimize the execution of single join operations for relations stored on tape. Careful selection of the processing block size and the ordering of block accesses is demonstrated to reduce execution time by about a factor of 10. 24] exploits the use of ....
S. Sarawagi and M. Stonebraker. "Single Query Optimization for Tertiary Memory," Proc. of the 1994 SIGMOD Conference, May, 1994.
No context found.
S. Sarawagi and M. Stonebraker, "Single Query Optimization for Tertiary Memory," Technical Report UCB:S2K-94-45, Univ. of California, Berkeley, 1994.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC