Chapter 22:Parallel And Distributed Query Processing ■Overview ■Parallel Sort ■Parallel Join ■Other Operations Parallel Evaluation of Query Plans Query Processing on Shared Memory Query Optimization Distributed Query Processing Database System Concepts-7th Edition 22.2 @Silberschatz,Korth and Sudarshan
Database System Concepts - 7 22.2 ©Silberschatz, Korth and Sudarshan th Edition Chapter 22: Parallel And Distributed Query Processing ▪ Overview ▪ Parallel Sort ▪ Parallel Join ▪ Other Operations ▪ Parallel Evaluation of Query Plans ▪ Query Processing on Shared Memory ▪ Query Optimization ▪ Distributed Query Processing
Parallel Query Processing Different queries/transactions can be run in parallel with each other. Interquery parallelism Concurrency control takes care of conflicts in case of updates More on parallel transaction processing in Chapter 23 Focus in this chapter is on read-only queries Individual relational operations (e.g.,sort,join,aggregation)can be executed in parallel data can be partitioned and each processor can work independently on its own partition. Queries are expressed in high level language (SQL,translated to relational algebra) makes parallelization easier. Database System Concepts-7th Edition 22.3 ©Silberscha乜，Korth and Sudarshan
Database System Concepts - 7 22.3 ©Silberschatz, Korth and Sudarshan th Edition Parallel Query Processing ▪ Different queries/transactions can be run in parallel with each other. • Interquery parallelism • Concurrency control takes care of conflicts in case of updates • More on parallel transaction processing in Chapter 23 • Focus in this chapter is on read-only queries ▪ Individual relational operations (e.g., sort, join, aggregation) can be executed in parallel • data can be partitioned and each processor can work independently on its own partition. ▪ Queries are expressed in high level language (SQL, translated to relational algebra) • makes parallelization easier
Intraquery Parallelism Intraquery parallelism:execution of a single query in parallel on multiple processors/disks;important for speeding up long-running queries. Two complementary forms of intraquery parallelism: Intraoperation Parallelism-parallelize the execution of each individual operation in the query Supports high degree of parallelism Interoperation Parallelism-execute the different operations in a query expression in parallel. Limited degree of parallelism Database System Concepts-7th Edition 22.4 ©Silberscha乜，Korth and Sudarshan
Database System Concepts - 7 22.4 ©Silberschatz, Korth and Sudarshan th Edition Intraquery Parallelism ▪ Intraquery parallelism: execution of a single query in parallel on multiple processors/disks; important for speeding up long-running queries. ▪ Two complementary forms of intraquery parallelism: • Intraoperation Parallelism– parallelize the execution of each individual operation in the query ▪ Supports high degree of parallelism • Interoperation Parallelism– execute the different operations in a query expression in parallel. ▪ Limited degree of parallelism
Parallel Processing of Relational Operations Our discussion of parallel algorithms assumes: ·read-only queries shared-nothing architecture ·n nodes,N1,,Nn Each assumed to have disks and processors. Initial focus on parallelization to a shared-nothing node Parallel processing within a shared memory/shared disk node discussed later Shared-nothing architectures can be efficiently simulated on shared- memory and shared-disk systems. Algorithms for shared-nothing systems can thus be run on shared-memory and shared-disk systems. However,some optimizations may be possible. Database System Concepts-7th Edition 22.5 ©Silberscha乜，Korth and Sudarshan
Database System Concepts - 7 22.5 ©Silberschatz, Korth and Sudarshan th Edition Parallel Processing of Relational Operations ▪ Our discussion of parallel algorithms assumes: • read-only queries • shared-nothing architecture • n nodes, N1 , ..., Nn ▪ Each assumed to have disks and processors. • Initial focus on parallelization to a shared-nothing node ▪ Parallel processing within a shared memory/shared disk node discussed later • Shared-nothing architectures can be efficiently simulated on sharedmemory and shared-disk systems. ▪ Algorithms for shared-nothing systems can thus be run on shared-memory and shared-disk systems. ▪ However, some optimizations may be possible
Intraoperation Parallelism Database System Concepts -7th Edition 22.6 @Silberschatz,Korth and Sudarshan
Database System Concepts - 7 22.6 ©Silberschatz, Korth and Sudarshan th Edition Intraoperation Parallelism