S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. ACM Trans. on Database Systems, 24(2):177--228, 1999.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....plan. Cost based optimizers only use the selectivity of the predicates to order them in the query plan but do not consider their computational complexities. In [49, 48] Hellerstein et al. use both the selectivity and the cost of selection predicates to optimize queries with expensive UDFs. In [26], Chaudhari and Shim proposes dynamic programming based algorithms to optimize queries with expensive predicates. The techniques discussed above deal with UDF optimization when the functions appear in the where clause of an SQL query. Expensive UDFs can also appear in the projection clause as ....

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. In Proc. 22nd Int. Conf. on Very Large Data Bases VLDB '96, pages 87--98, 1996.

....system of the XQuery core significantly. However, the XQuery core cannot properly type recursive XML queries [2, 10, 11] In this regard, our structural function inlining is a novel technique for typing recursive XML queries. As to optimizing functions, most of existing optimization techniques [6, 7] treat functions simply as externally defined black boxes accompanying some semantic information. Moreover, they consider nonrecursive functions only, and even the XQuery core cannot optimize recursive functions [2, 10, 11] In contrast, the structural function inlining optimizes recursive ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. ACM Transactions on Database Systems, 24(2):177-- 228, June 1999.

....Here the cost of the plan includes the join cost and the evaluation costs of all complex operators. The final join plan for our example is shown on the fight hand side of Figure 8. Notice that our algorithm is not exhaustive in terms of possible alternatives for complex operator placement (as is [CS96] for predicates) This is an intentional compromise done to avoid the extra combinatorial explosion of such an exhaustive search. At present, we have not completed the implementation of the cost based query optimizer for the QPC although the major building blocks, such as query plans and search ....

S. Chaudhuri and K. Shim. Optimization of Queries with User-defined Predicates. In Proc. 22nd VLDB Conf., pp. 87 98, Bombay, India, 1996.

....and cost of user defined functions is non trivial, although a reasonable guess may be accomplished by maintaining statistics of previous executions. Some researchers have proposed that the placement of expensive predicates such as UDFs be considered during the join enumeration algorithm [6, 20]. This approach has the e#ect of substantially increasing the cost of enumeration. Hellerstein proposed a predicate migration approach [20] which can lead to cost that is polynomial in the number of UDFs. In the worst case, predicate migration requires exhaustive enumeration of the join space, ....

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. In Proc. Int'l Conf. on VLDB, Mumbai(Bombay), India, 1996.

....and partitioning is beyond the scope of this paper. In the following, we will describe changes to the optimizer s cost model and the plan enumerator. These changes are essentially along the lines of previous work to extend bottom up, dynamic programming query optimizers, e.g. Loh88,CS94,CS96,CK97] Cost Model The cost model extensions are straightforward. We only need to provide cost estimates for all new operators like OHJ, SOHJ, indirect partitioning, and BulkMerge. Similar cost formulae as those needed to estimate the cost of (S)OHJ operators have been devised in [BCK98,BCKK00] ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 87--98, Bombay, India, September 1996.

....is read (i.e. at which cycle provider and with which scan or wrapper operator) After that, join plans are constructed from these access plans and (later) from simpler join plans. To deal with unary external functions and predicates, the dynamic programming algorithm is extended as described in [8]. In every step, the cost of each plan is estimated and inferior plans are pruned in order to speed up the optimization process. Rather than presenting the full details of the ObjectGlobe optimizer, we would like to highlight the peculiarities that make the ObjectGlobe optimizer special: ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 87--98, Bombay, India, September 1996.

....is read (i.e. at which cycle provider and with which scan or wrapper operator) After that, join plans are constructed from these access plans and (later) from simpler join plans. To deal with unary external functions and predicates, the dynamic programming algorithm is extended as described in [CS96] In every step, the cost of each plan is estimated and inferior plans are pruned in order to speed up the optimization process. Rather than presenting the full details of the ObjectGlobe optimizer, we would like to highlight the peculiarities that make the ObjectGlobe optimizer special: ....

....standard exception handling mechanism of Java with a little help from our send receive operator pair for crossing network connections. The servers of child operators cannot be informed with the exception mechanism. A 14 Total Lookup Time Avg. Time per Search Optimization Time Scenario I 5. 64 secs 0.47 secs 0.83 secs Scenario II 5.64 secs 0.47 secs 0.07 secs Table 1: Overheads of Plan Generation wrapper execution location time Passau 151 secs Maryland 62 secs Table 2: Query for all Hotels in Philadelphia special (UDP) network protocol is used for this purpose. What we did not mention ....

[Article contains additional citation context not shown here]

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In VLDB [VLD96], pages 87--98.

....about: ffl Kim [Kim82] ffl Ganski Wong [GW87] ffl Dayal [Day87] ffl Magic Sets [MP92, SPL96, SHP 96b] ffl Cluet and Moerkotte [CM95, CM93] ffl Steenhagen [SABd94] 7. 4 Semantic Optimization and Handling Foreign Functions ffl Zdonik [HZ80] ffl Aberer [AF95] ffl Chaudhuri and Kim [CS96, CS93] ffl Conjunctive Predicates [HS93, Hel94, LMS94] ffl Avoiding sorting [SSM96] ffl Semantics and Cost Estimation [NCN97] 7.5 Plan Languages, Partial Evaluation and Dynamic Optimization ffl Exodus Volcano [CDG 90, GM93] ffl OPA [DGK 91, Gra95] ffl Opt [KD] ffl ....

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. In Proceedings of the 22nd VLDB Conference, pages 87--110, Bombay, India, September 1996.

....is read (i.e. at which cycle provider and with which scan or wrapper operator) After that, join plans are constructed from these access plans and (later) from simpler join plans. To deal with unary external functions and predicates, the dynamic programming algorithm is extended as described in [CS96] In every step, the cost of each plan is estimated and inferior plans are pruned in order to speed up the optimization process. Rather than presenting the full details of the ObjectGlobe optimizer, we would like to highlight the peculiarities that make the ObjectGlobe optimizer special: 10 ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 87--98, Bombay, India, September 1996.

....early sorting and partitioning is beyond the scope of this paper. In the following, we will describe changes to the optimizer s cost model and the plan enumerator. These changes are essentially along the lines of previous work to extend bottom up, dynamicprogramming query optimizers, e.g. [Loh88,CS94, CS96,CK97]. Cost Model The cost model extensions are straightforward. We only need to provide cost estimates for all new operators like OHJ, SOHJ, indirect partitioning, and BulkMerge. Similar cost formulae as those needed to estimate the cost of (S)OHJ operators have been devised in [BCK98, BCKK00] for ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. of the Conf. on Very Large Data Bases (VLDB), pages 87--98, Bombay, India, September 1996.

....Here the cost of the plan includes the join cost and the evaluation costs of all complex operators. The final join plan for our example is shown on the right hand side of Figure 8. Notice that our algorithm is not exhaustive in terms of possible alternatives for complex operator placement (as is [CS96] for predicates) This is an intentional compromise done to avoid the extra combinatorial explosion of such an exhaustive search. At present, we have not completed the implementation of the cost based query optimizer for the QPC although the major building blocks, such as query plans and search ....

S. Chaudhuri and K. Shim. Optimization of Queries with User-defined Predicates. In Proc. 22nd VLDB Conf., pp. 87-- 98, Bombay, India, 1996.

....operators in step (9) These are operators whose arguments come from more than one relation. The final join plan for our example is shown on the right hand side of Figure 9. Notice that our algorithm is not exhaustive in terms of possible alternatives for complex operator placement (as is [CS96] for predicates) This is an intentional compromise done to avoid the extra combinatorial explosion of such an exhaustive search. At present, we have not completed the implementation of the cost based query optimizer for the QPC although the major building blocks, such as query plans and search ....

....search. At present, we have not completed the implementation of the cost based query optimizer for the QPC although the major building blocks, such as query plans and search procedures, are in place. In MOCHA, not only do we have 17 to deal with the placement of complex predicates [HS93, CS96] but also deal with complex projections and aggregates as well, thus making the optimization process more complicated. We are exploring a series of pruning heuristics to reduce the search space of the optimizer, and speed up the optimization process. 5 Performance Evaluation To validate our ....

S. Chaudhuri and K. Shim. Optimization of Queries with User-defined Predicates. In Proc. 22nd VLDB Conference, pages 87--98, Bombay, India, 1996.

....to enable decision support. Optimization concerns the ways in which factual data can be efficiently stored and manipulated. There are two main approaches to this problem in the context of decision support applications: materialization of pre computed summary data [13,21] and query optimization [3,7]. The formal nature of the model proposed here is well suited for an investigation of the above problems. In particular, we are currently developing an algebra for the MultiDimensional model, for studying the efficient evaluation of multidimensional queries. Acknowledments We would like to ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Twenty-second Int. Conf. on Very Large Data Bases, Bombay, pages 87--98, 1996.

....applied to the multi operator implementations in the usual way to speed up processing further, i.e. by means of data parallelism and pipelining. 6. Related work User Defined Functions (UDFs) have attracted increasing interest of researchers as well as the industry in recent years (see e.g. 1] [7], 15] 16] 24] 27] 32] 34] 38] 40] However, most of the work discusses only the non parallel execution of UDFs, special implementation techniques like caching, or it is directed towards query optimization for UDFs. In [34] pipeline parallelism for functions as well as ....

Chaudhuri, S., Shim, K.: Optimization of Queries with User-defined Predicates. VLDB Conf. 1996: 87-98.

....intermediate results can have exponential size, as noted in Example 7. In fact, no reorderings are saved in Example 7 at all, unless the resulting disjuncts are reoredered to apply the string formulae OE i last. This problem is similar to optimizing left deep joins with user defined predicates [2], except that our predicates not only test existing relation columns but also create new ones, and their cost is heavily (even exponentially) influenced by the number of created columns. In other words, we are adding user defined set valued functions, whose costs depend on the expected sizes of ....

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Very Large Data Bases Conference, pages 87--98, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In VLDB'96, Proceedings of 22th International Conference on Very Large Data Bases, September 3-6, 1996.

....is to determine the set of filter conditions that are to be evaluated using GradeSearch. The rest of the conditions will be evaluated by using Probe. In order to efficiently execute the latter step, we will exploit the known techniques in optimizing the processing of expensive filter conditions [25, 22, 23, 26, 11]. In this section, we first define a space of search minimal executions, which access as few attributes as possible using GradeSearch, and sketch the cost model and the optimization criteria. Next, we describe an optimization algorithm and explain the conditions under which it is optimal. ....

....objects in the repository, IOal Sel(a) o. Optimizing Evaluation of Residues: Given a residue (a, f) the task of determining an optimal eval uation for (a, f) maps to the well studied problem of optimizing the execution of selection conditions containing expensive predicates [25] See also [23, 26, 22, 11]. If (a, f) is a conjunction of atomic conditions a A. A a, there is an efficient algorithm w that finds the optimum probing strategy. Specifically, it can be shown [23, 26] that the order in which the atomic conditions for each object should be probed is given by the rank of each condition ....

[Article contains additional citation context not shown here]

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. ACM Transactions on Database Systems, 24(2): 177-228, June 1999.

.... more general than a conjunction of literals, and (b) an expensive predicate can be evaluated by using GradeSearch (indexed access) as well as by using Probe (both the arguments of the predicate are bound) Although past work on expensive predicates addresses the case where only condition (b) holds [13, 14], it does not address the case where condition (a) holds as well. Finally, note that the algorithms and results of the previous section are completely independent of the nature of the atomic predicates as long as a selectivity and a cost measure are available. Several interesting query processing ....

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. In Proceedings of the 22nd VLDB Conference, Bombay, India, September 1996.

....are no joins, then the expensive predicates may be ordered efficiently by their ranks, computed from their selectivity and per tuple cost of evaluation. Unfortunately, their attempt to extend the use of ranks for queries with joins may result in suboptimal plans. This shortcoming is resolved in [8] by representing the application of user defined predicates like a physical property of a plan so that the dynamicprogramming based enumeration algorithm guarantees optimality. Moreover, for realistic assumptions of the cost model, it is shown that the problem is polynomial in the number of ....

Chaudhuri, S., Shim K. Optimization of Queries with Userdefined Predicates. In Proc. of VLDB, Mumbai, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. ACM Trans. on Database Systems, 24(2):177--228, 1999.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. Int. Conf. on Very Large Data Bases (VLDB), pages 87--98, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. Technical report, Microsoft Research, Advanced Technology Division, One Microsoft Way, Redmond, WA 98052, USA, 1997.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In Proc. Int. Conf. on Very Large Data Bases (VLDB), pages 87--98, Bombay, India, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. ACM Trans. on Database Systems, 24(2):177--228, 1999.

No context found.

S. Chaudhuri and K. Shim. Optimization of Queries with User-defined Predicates. In VLDB, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In VLDB, pages 87--98, 1996.

No context found.

S. Chaudhuri and K. Shim. Optimization of queries with user-defined predicates. In VLDB, pages 87--98, 1996.

No context found.

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. ACM Transactions on Database Systems (TODS), 24(2):177--228, 1999.

No context found.

Chaudhuri, S., and K. Shim, Optimization of Queries with User-Defined Predicates. Proc. VLDB, 1996, 87-98.

No context found.

S. Chaudhuri and K. Shim. Optimization of Queries with User-defined Predicates. In Proc. of the 22nd VLDB Conference, pp. 87--98, Bombay, 1996.

No context found.

Chaudhuri, S. and Shim, K. "Optimization of Queries with User-defined Predicates" VLDB, 1996.

No context found.

Surajit Chaudhuri and Kyuseok Shim. Optimization of queries with user-defined predicates. In Proceedings of the Twenty Second International Conference on Very Large Databases (VLDB), Bombay, India, pages 87--98, September 1996.

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