Two notions of Parallel query processing
In Distributed Query Processing, the results are all stored in a single “head” node. In Parallel Query Processing, the results are stored across many nodes.
In Distributed Query Processing, queries are processed in parallel and the results are stored in a single “head” node. In Parallel Query Processing, queries are processed in parallel in parallel and stored among many nodes.
Distributed Query
- Rewrite the query as a union of subqueries
- Workers communicate through standard interfaces, so compatible with federated, heterogeneous, or distributed databases
Parallel Query
- Each operator is implemented with a parallel algorithm. See Teradata