Disk allocation methods for binary Cartesian product files

Abstract

Abstract.Since a file is usually stored on a disk, the response time to a query is dominated by the disk access time. In order to reduce the disk access time, a file can be stored on several independently accessible disks. In this paper, we discuss the problem of allocating buckets in a file among disks such that the maximum disk accessing concurrency can be achieved. We are particularly concerned with the disk allocation problem for binary Cartesian product files. A new allocation method is first proposed for the cases when the number (m) of available disks is a power of 2. Then it is extended to fit the cases where m is not a power of 2. The proposed algorithm has a "near" strict optimal performance for a partial match query in which the number of unspecified attributes is greater than a~ small number (5 or 6).

A greedy relaxed generalized disc modulo allocation method for multiple key hashing files on multi‐disc systems

“…Though the experimental solutions presented in [7] showed that Du's method can achieve near optimal performance for some PMQs, Du's method is not general enough in practice owing to the restriction that his algorithm can only be applied to Binary Cartesian Product file systems. Therefore, we will propose a more generalized and efficient algorithm based on the GDM allocation method to allocate MKH file systems onto multiple discs.…”

“…In 1986, Du [7] proposed a heuristic algorithm for the cases when the number of available discs is a power of two. He also showed that the proposed algorithm has a "near strictly optimal" performance for PMQs in which the number of unspecified attributes is greater than a small number (five or six) and gave some suggested criteria to modify the heuristic method such that it is suitable to those cases when ND is not equal to a power of two.…”

Optimality criteria of the disk modulo allocation method for Cartesian product files

“…By using their formula, Chang and Chen [1 I] showed that it is strictly optimal Whenever k = m, the number of available disks. In addition, the multidisk file allocation for an important special k-ary CP file with k = 2 (also called the binary CP file) has also been explored by many researchers in recent years [2,3,8,10,16,23].…”

Partitioning similarity graphs: A framework for declustering problems

“…al [7,18,11,351. We note that no methods can achieve strict optimality for all query-sets [7,18,351. For example, no method can be strictly optimal for all range queries if the number of disks is greater than 5 [18, 351.…”