Query processing in main memory database management systems

Abstract

Most previous work in the area of main memory database systems has focused on the problem of developing query processing techniques that work well with a very large buffer pool. In this paper, we address query processing issues for memory resident relational databases, an environment with a very different set of costs and priorities. We present an architecture for a main memory DBMS, discussing the ways in which a memory resident database differs from a disk-based database. We then address the problem of processing relational queries in this architecture, considering alternative algorithms for selection, projection, and join operations and studying their performance. We show that a new index structure, the T Tree, works well for selection and join processing in memory resident databases. We also show that hashing methods work well for processing projections and joins, and that an old join method, sort-merge, still has a place in main memory.

In-memory hash tables for accumulating text vocabularies

“…T-trees have been specifically proposed as a suitable structure for in-memory management of large sets of distinct search terms [3]. T-trees are similar to BSTs, with the modification that each node contains an array of lexicographically adjacent search terms, up to a fixed limit.…”

Optimizing database architecture for the new bottleneck: memory access

“…For higher selectivities, Lehman and Carey [LC86] concluded that the T-tree and bucket-chained hashtable were the best data structures for accelerating selections in main-memory databases. The work in [Ron98] reports, however, that a B-tree with a block-size equal to the cache line size is optimal.…”

Efficiently instantiating view-objects from remote relational databases

“…In main memory databases, the choice of query processing strategies (DeWitt et al, 1984, Bitton andTurbyfill, 1986;Shapiro, 1986;Lehman and Carey, 1986b;Bitton et al, 1987;Swami, 1989;Whang and Krishnamurthy, 1990) is based on the number of CPU cycles and memory space efficiency rather than the number of disk accesses and disk space efficiency. The results of comparing different query processing strategies obtained by some researchers (DeWitt et al, 1984;Lehman and Carey, 1986b;Shapiro, 1986) showed that hash-based query processing strategies are faster than others when large main memory is available.…”

“…The results of comparing different query processing strategies obtained by some researchers (DeWitt et al, 1984;Lehman and Carey, 1986b;Shapiro, 1986) showed that hash-based query processing strategies are faster than others when large main memory is available. On the other hand, a main memory database system used in OBE (Bitton et al, 1987;Whang et al, 1987;Whang and Krishnamurthy, 1990) implemented apipelined nested loop join with array indexes and obtained good performance in both time and memory space.…”

“…lndex Creation. Once redundant joins are removed, indexes are created on the join attribute of each RF except the pivot RE According to the performance study by Lehman and Carey (1986a) a chained bucket hashing gives the best performance among all main memory index operations except for a range query. Because we do not need a range query, bucket hashing is appropriate for our use.…”