Query Processing in a Symmetric Parallel Environment
Query Processing in a Symmetric Parallel Environment
Dennis Shasha
The book Query Processing in a Symmetric Parallel Environment was written by author Dennis Shasha Here you can read free online of Query Processing in a Symmetric Parallel Environment book, rate and share your impressions in comments. If you don't know what to write, just answer the question: Why is Query Processing in a Symmetric Parallel Environment a good or bad book?
What reading level is Query Processing in a Symmetric Parallel Environment book?
To quickly assess the difficulty of the text, read a short excerpt:
Moreover, since the query gfaph has no self-loops (by our preprocessing) R # S. Therefore, \^fni OT without (R. X, S. Y}, R. X will be in a clause associated with R and similarly^for S. Y. So, the nodecost of R and S are the same in the new graph and th6 did bi5e7 □ Corollary: Any two equivalent queries have the same repartitioning cost. n Suppose q is a 'single-clause tree query. By proposition 1, QG(^'^) is singleton. By propositjon 3, every spanning tree of QG(^"^) is equivalent to q. Hence ...every spanning tree, of QG(q) is also equivalent to q. By proposition 4, every such spanning tree has the same cost. Given a query q that we suspect is a single-clause tree query, we do the following single-clause-optimize(q: query) begin construct a query graph for q; eliminate self -loops; take any spanning tree QG(q') of QG(q); if q' is equivalent to q then {q is a single-clause tree query} {q' is an optimal repartitioning spanning tree} return q' else {q is not a single-clause tree query} try other heuristics on q end if end We thus have a constructive proof of the following theorem.
User Reviews: