G. Graefe and W. McKenna, "The Volcano optimizer generator: Extensibility and efficient search", Proc. of 9th IEEE Intl. Conf. on Data Engineering (ICDE), April 1993.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....costing is done, the two algorithms perform exactly the same steps to find the optimal plan. 3.2. Exhaustive with Exact Pruning An optimizer may be able to save a considerable amount of computation by pruning away subplans that it knows will not be part of any optimal plan. A top down approach [21, 20] is more suitable for this kind of pruning than the bottom up dynamic programming approach we described above, though it is possible to incorporate pruning in that algorithm as well. Typically, these algorithms first find some plan for the query and then use the cost of this plan to prune away ....

G. Graefe and W. J. McKenna. The volcano optimizer generator: Extensibility and efficient search. In ICDE, 1993.

....them. That is the case for the arrable Trades in that figure. We will show shortly how arrables having vector elements, such as the arrable Series in the same figure, can be useful. 05 11 03 05 11 03 05 11 03 05 11 03 12.02 43.23 12.04 12.05 43.22 price tradeDate Series ACME WXYZ [1 5 9] [2 13] ID ts tradeDate ID [12.02 12.04 12.05] 43.23 43.22] ACME WXYZ ACME ACME WXYZ 1 2 5 9 13 Figure 1: Example of two well formed arrables Definition 2 (Arrable Indexing) The k th record of an arrable r is formed by the k th element of each of r s component arrays. This ....

....Horizontal axis is the number of hosts for (a) and (b) over which the 1 million connections are divided and the number of securities for (c) and (d) over which the 1 million trades are divided. Plans for (a) are shown in figure 3, for (b) in 4, for (c) in 5, and for (d) in 6. enforcer [5] made sure a sort step was added whenever an e#cient algorithm required it. In [17] the authors added an order optimization step before plans were enumerated in the context of DB2 s optimization process. This step may dramatically improve queries that have order requirements due to the clause ....

Graefe, G., and McKenna, W.J., "The Volcano Optimizer Generator: Extensibility and E#cient Search." ICDE Int'l Conf on Data Engineering, 209--218, 1993.

....equivalences or transformation rules that let it consider alternative plans. To attack this problem we propose a number of transformation rules that modify query plan trees containing the GApply operator. This allows the GApply operator to be seamlessly integrated into a Volcano style optimizer [13]. We note here that since GApply has been identified to be useful in the data warehousing context, all our rules are automatically applicable to decision support queries. 3. Finally, based on our experiments with SQL Server 2000, we argue that it is necessary to expose GApply in the syntax since ....

G. Graefe and W. McKenna. The volcano optimizer generator: Extensibility and e#cient search. In ICDE, 1993.

....[3, 4] was developed as a testbed based on the VODAK OODBMS [19] Extending existent facilities of the underlying OODBMS for modeling and querying structured documents was identified and adopted at the inception of our research. Extensibility ever motivated intensive research for extensible DBMS [10]. Our experience with structureddocument query optimization reveals that a plain application of the extensible DBMS approach does not work well in practice as expected. In our case, we extended the VODAK query optimizer by adding quite a number of new transformation rules exploiting the ....

G. Grafe and W.J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proc. of 9th ICDE, pp. 209-218, Vienna, Austria, April 19-23, 1993.

....techniques or search strategies have to be changed [Cha98] As a result, the last decade has witnessed substantial efforts aiming to develop extensible query optimizers that would make such changes easier. Representative examples of extensible query optimizers include Starburst [Haa90] Volcano [GM93], and OPT [KW99] This paper reports on a specific study that has enhances the Volcano extensible query optimizer to support a relational algebra with temporal operators such as temporal join and aggregation. In addition to new operators, cost formulas, selectivity estimation formulas, and ....

....improved performance. The search strategy of Volcano is fixed, and no mechanisms for extending or changing it are provided. Proposed improvements of Volcano that were not part of the available code include a mechanism for heuristic guidance, where rules can be ordered according to their promise [GM93]. Such ordering implies that the rules having the best probability to yield better plans would be applied as soon as possible, reducing the overall plan search time. We had to add support for equivalence class elements that point to their own equivalence classes, because this facility was not ....

G. Graefe and W. J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proceedings of IEEE ICDE, Vienna, Austria, pp. 209--218 (1993).

....existing optimization algorithms to use cost functions in place of costs. We show how to extend the System R query optimization algorithm [SAC 79] to perform parametric query optimization with piecewise linear cost functions. We have also extended the Volcano query optimization algorithm [GM93] in a similar fashion. The solution works for an arbitrary number of parameters. The rest of the paper is organized as follows. Section 2 formally defines the parametric query optimization problem and provides background material on polytopes. Section 3 describes non intrusive algorithms for PQO ....

Goetz Graefe and William J. McKenna. The Volcano optimizer generator: Extensibility and e#cient search. In Proc. of the ICDE, pages 209--218, 1993.

....system. It often implements a rule rewriter to perform equivalent transformation on the query expressions. Representatives of this kind of system include: the System R style Optimizer [15] Starburst project [16] 31] the Exodus Optimizer Generator [7] and the Volcano Optimizer Generator [8]. The system R style Optimizer designs sets of rules to translate a query into a physical plan. One set of them is to convert the query into an algebraic tree. Other sets are used to generate access paths, join orders, and join methods. The optimizer developed in the Starburst project rst uses a ....

G. Graefe and W.J. McKenna. The Volcano Optimizer Generator: Extensibility and Ecient Search. Proceedings of IEEE Conference on Data Engineering, Vienna, Austria, 1993. 174

....are derived from OPERATOR, representing a logical and physical operator respectively, and contain a pointer to one another. The spine of the operator tree is maintained within the logical operator. As studied in the Volcano and OPT efforts, this separation of algebra has proven effective [48,57]. The non shaded area of Figure 22 illustrates the classes implemented for the VenusDB optimizer. Recall, VenusDB rules are evaluated using nested loops. Each loop ranges over a container with either an existential or universal cursor (Sections 3.1 and 5.2.1.1) Thus, the VenusDB operator tree ....

....to sequence and prune the search space. 5.2.2.3.1 Rewrite System The search space is defined by the space of algebraic rewrites presented in the rewrite system. Like the previous generation of rule based optimizers, the rewrite system in the Venus based optimizer is implemented in rule based form [32,47,48,80]. The rewrite system applies different transformation and implementation rewrites on a subquery by exploiting procedural and heuristic elements, constrained by the algebraic representation. If the preconditions and conditions for a rewrite are satisfied, the operator is applied and the new ....

G. Graefe and W. McKenna, "The Volcano optimizer generator: Extensibility and efficient search," in Proceeding of the 12th International Conference on Data Engineering. Vienna, Austria, April, 1993, 209-218.

....externally determined thresholds. Furthermore, our upper bounds can differ for each subplan being optimized. Top down optimization began with the Exodus optimizer generator [GrD87] whose primary purpose was to demonstrate extensibility. Graefe and collaborators subsequently developed Volcano [GrM93] with the primary goal of improving efficiency with memoization. Volcanos efficiency was hampered by its search strategy, which generated all logical expressions before generating any physical expressions. This ordering meant that Volcano generated O (3 N ) expressions, like Starburst. ....

G. Graefe and W. J. McKenna, The Volcano Optimizer Generator: Extensibility and Efficient Search, Proc. Data Engineering Conf. 1993, Pg. 209-218.

No context found.

G. Graefe and W. McKenna, "The Volcano optimizer generator: Extensibility and efficient search", Proc. of 9th IEEE Intl. Conf. on Data Engineering (ICDE), April 1993.

No context found.

G. Graefe and W. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In ICDE, 1993.

No context found.

G. Graefe and W. J. McKenna. The volcano optimizer generator: Extensibility and efficient search. In ICDE, pages 209--218, 1993.

No context found.

G. Graefe and W. J. McKenna. The volcano optimizer generator: Extensibility and efficient search. In ICDE, pages 209--218, 1993.

No context found.

G. Graefe and W. McKenna. The volcano optimizer generator: Extensibility and efficient search. In Proceedings of ICDE, pages 209--218, 1993.

No context found.

G. Graefe and W. J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proceedings of IEEE ICDE, pp. 209--218 (1993).

No context found.

G. Graefe and W. J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proceedings of IEEE ICDE, Vienna, Austria, pp. 209--218 (1993).

No context found.

Goetz Graefe and William J. McKenna. The Volcano optimizer generator: Extensibility and e#- cient search. In ICDE, 1993.

No context found.

G. Graefe and W. J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proc. IEEE ICDE, pp. 209--218 (1993).

No context found.

Goetz Graefe and William J. McKenna. The volcano optimizer generator: Extensibility and e#cient search. In Proceedings of the Ninth International Conference on Data Engineering, April 19-23, 1993, Vienna, Austria, pages 209--218. IEEE Computer

No context found.

G. Graefe and W. J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proc. of the IEEE Int'l. Conf. on Data Engineering, pages 209--218, Vienna, Austria, Apr. 1993.

No context found.

Goetz Graefe and William J. McKenna. The volcano optimizer generator: Extensibility and efficient search. In Proceedings of the Ninth International Conference on Data Engineering, April 19-23, 1993.

No context found.

Goetz Graefe and W. McKenna. The Volcano Optimizer Generator: Extensibility and Ecient Search. In IEEE Conference of Data Engineering, Wien, 1993.

No context found.

G. Graefe and W. J. McKenna. The volcano optimizer generator: Extensibility and efficient search. In Proceedings of the Ninth International Conference on Data Engineering, April 19-23, 1993.

No context found.

G. Graefe and W.J. McKenna. The Volcano Optimizer Generator: Extensibility and Efficient Search. In Proc. of the 9th IEEE Intl. Conf. on Data Engineering, pp. 209-- 218, Vienna, 1993.

No context found.

G. Graefe and W. McKenna. The volcano optimizer generator: Extensibility and efficient search. In Proceedings of ICDE, pages 209--218, 1993.

First 50 documents  Next 50

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC