Parallel Query Processing Techniques In Parallel Database System PBASE/2

This dissertation addresses parallel query processing techniques in a parallel database system named PBASE/2, which is designed and implented all by ourselves.This dissertation is made up of three parts.In part one, I survey the overall aspects of PBASE/2. Firstly, a abstract parallel computing model APCM based on shared-nothing architecture is put forward. Then the overall system architecture of PBASE/2 is described. Finally, I will introduce the parallel query evaluation mode of PBASE/2, which is the basis and springboard of parallel query processing.Parallel query processing consists of two not only independent but also correlative portions — parallel query evaluation and parallel query optimization. Part two of this dissertation introduces the parallel query evaluation engine of PBASE/2, that is, a virtual database machine. Virtual database machine is a software abstract layer on general computing platforms, especially parallel computing platforms, and is made up of three components: virtual processor, virtual data bus and virtual memory. Virtual processors manage all CPU resources in the system, on which various operations are running. Virtual data bus is responsible for the management of data transmission between associated operations, which forms the hinges of the entire system. Virtual memory provides efficient data storage and buffering mechanisms that conform to data reference behaviors in database systems. The architecture of virtual database machine is very clear and has many good features such as high efficiency, high scalability, high extensibility, high portability etc. virtual database machine provides an excellent parallel query evaluation engine for PBASE/2.Part three addresses another important portion of parallel query processing—parallel query optimization. For the sake of cutting down the enormous search space of parallel query optimization, PBASE/2 adopts a...