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.
􀃌 Interoperation Parallelism – execute the different operations in a
query expression in parallel.
the first form scales better with increasing parallelism because
the number of tuples processed by each operation is typically
more than the number of operations in a query


Parallel Processing of Relational Operations
􀂄 Our discussion of parallel algorithms assumes:
􀃌 read-only queries
􀃌 shared-nothing architecture
􀃌 n processors, P0, ..., Pn-1, and n disks D0, ..., Dn-1, where disk Di is
associated with processor Pi.
􀂄 If a processor has multiple disks they can simply simulate a
single disk Di.
􀂄 Shared-nothing architectures can be efficiently simulated on
shared-memory and shared-disk systems.
􀃌 Algorithms for shared-nothing systems can thus be run on sharedmemory
and shared-disk systems.
􀃌 However, some optimizations may be possible.




Comments

Popular posts from this blog

Handling of Skew

Fragment-and-Replicate Join

USER INTERFACE DESIGN FOR ANNA UNIVERSITY SYLLABUS