Mapping a parallel complex-object DBMS to operating system processes

Abstract

So far, parallelism in complex-object and object-oriented DBMS has not been investigated in depth because descriptive and set-oriented query languages did not exist for these systems. However, with standardization of OQL by ODMG or SQL3 by ANSI, systems implementing these languages are ready to exploit parallel query processing strategies. In this paper, we explain differences between parallel query processing in relational and complex-object DBMS. Furthermore, we present a client~server-based system architecture that allows for fine-grained parallelism within query processing in complex-object DBMS. We investigate various strategies for mapping this architecture to processes of the underlying operating system. Finally, some measurements show the impact of these mapping strategies on query response times.

Measuring and modelling the performance of a parallel ODMG compliant object database server

“…Scientific parallel database systems, such as Volcano [38], MIDAS [39], XPRS [40], Gamma [23], Monet [32], Bubba [15], Cocoon [41], Goblin [42] and Prima [18], provide a declarative interface for querying data, which allows users to be insulated from details on the machine architecture and physical characteristics of data. In those systems, query optimization and parallelization are automatic, having as a result an operator tree that represents a query's evaluation plan.…”

“…However, the language on which this is based [16] seems less powerful than OQL, and the query processing framework is significantly different; it is not obvious to us that it can be adapted easily for use with OQL. Another parallel object-based database system is PRIMA [18], which uses the MAD data model and a SQL-like query language, MQL. The PRIMA system's architecture differs considerably from Polar's, as it is implemented as a multiple-level client-server architecture, where parallelism is exploited by partitioning the work associated with a query into a number of service requests that are propagated through the layers of the architecture in the form of client and sever processes.…”

A Cost Model for Parallel Navigational Access in Complex-Object DBMSs

“…Therefore, integrating parallel strategies into query processing engines seems to be a promising idea for achieving short query response times in OODBMSs, too. Unfortunately, due to different query processing characteristics [6] and access operations [8,10], well known parallel query processing techniques of relational systems often cannot be applied in OODBMSs. Especially, navigational access from objects to referenced objects which is a key property in OODBMSs and complex-object DBMSs is not supported in relational systems.…”