M. Garofalakis and Y. Ioannidis. Parallel query scheduling and optimization with time- and space-shared resources. In VLDB, 1997.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....two sites involved in the query. No Pipelining Across Sites : We assume that there is no pipelining of data among query operators across sites. The main issue with pipelining across sites is that the pipelined operators tend to waste resources, especially space shared resources such as memory [19]. Even if the producer is not slow, the communication link between the two sites could be slow, especially for WANs, and the consumer will be holding resources while waiting for the network. 3. Adapting the Optimization Techniques In this section, we discuss our adaptations of various well known ....

....the total communication cost may not necessarily decrease. Plan Space : 33] discusses the plan space explored by this algorithm. It will be a subspace of the plan space explored by the exhaustive algorithm. 3.4. Two phase Optimization Two phase optimization [26] has been used extensively [9, 19] in distributed and parallel query optimization mainly because of its simplicity and the ease of implementation. This algorithm works in two phases : The techniques described in [33] based on minimum selectivity, etc. can be applied orthogonally. However, since they do nothing to reflect ....

M. N. Garofalakis and Y. E. Ioannidis. Parallel query scheduling and optimization with time- and space-shared resources. In VLDB, 1997.

....data delivery rate from multiple sources, as well as to deal with varying computing and I O resources. Although their focus is different than ours, it shows possible ways of incorporating low level requirements in terms of available memory in our scheduling algorithms. Garofalakis and Ioannidis [10] present a model to handle query scheduling on hierarchical parallel systems. This work is interesting in that it is one of the first that we are aware of that handles the complexities of modeling the multidimensional aspect of the multiple resources needed to compute relational queries. 16 7 ....

Minos N. Garofalakis and Yannis E. Ioannidis. Parallel query scheduling and optimization with time- and spaceshared resources. In Proceedings of the 23th VLDB Conference, pages 296--305, Athens, Greece, 1997.

....is essentially the best possible unless NP=ZPP. In addition to their theoretical importance, these problems have several applications such as load balancing, cutting stock, and resource allocation, to name a few. One of our motivations for studying these problems comes from recent interest [12, 13, 14] in multi dimensional resource scheduling problems in parallel query optimization. A favored architecture for parallel databases is the so called shared nothing environment [5] where the parallel system consists of a set of independent processing units each of which has a set of time sharable ....

....by a single aggregate work measure. This simplification is done typically to reduce the complexity of the scheduling problem. However, for large task systems that are typically encountered in database applications, ignoring the multi dimensionality could lead to bad performance. The work in [11, 12, 13, 14] demonstrates the practical effectiveness of the multi dimensional approach. One of the basic resource scheduling problem that is considered in the above papers is the problem of scheduling d dimensional vectors (tasks) on d dimensional bins (machines) to minimize the maximum load on any dimension ....

[Article contains additional citation context not shown here]

Minos N. Garofalakis and Yannis E. Ioannidis. Parallel query scheduling and optimization with time-and space-shared resources. In Proceedings of the 23rd VLDB Conference, pages 296--305, 1997.

....into sub plans which are then executed in parallel on di erent nodes of the parallel processing environment. The granularity of this decomposition varies and can be as ne as parallelizing single operator as studied for example in [SD89, SD90, WFA95] but is often chosen coarser [HM95, CHM95, GI96, GI97] 3. Architecture 3 These approaches have in common that they require communication between single nodes for shipping or exchanging partial results. This causes network contention and synchronization e ects where nodes have to wait for others to complete their tasks rst. As a result, a ....

M. N. Garofalakis and Y. E. Ioannidis. Parallel Query Scheduling and Optimization with Time- and Space-Shared Resources. In Proc. of the Int'l. Conf. on Very Large Data Bases, pages 296-305, Athens, Greece, September 1997.

....skew handling as well as intra operator parallelism. Chekuri et al. develop a more general treatment and allow for arbitrary query plans using the same cost model [2] Again, skew is not considered. Garofalakis and Ioannidis discuss a richer cost model and focus on shared nothing architectures [4, 5]. Their scheduling heuristics are also based on the assumptions that no skew affects the execution. Lo et al. study constraint processor allocation for pipelined hash joins [13, 3] and extend this approach in [10] to scheduling of separate pipelining segments. For their simulation model, they ....

M. N. Garofalakis and Y. E. Ioannidis. Parallel Query Scheduling and Optimization with Time- and Space-Shared Resources. In Proc. Int'l. Conf. on Very Large Data Bases, Athens, Greece, September 1997.

....25, 8] A version closely related to ours is the one in which there is a bound on the number of jobs that may be in the shop at any time [20] our formulation is more general. Although many results exist on scheduling parallel processors and database queries with additional resource constraints [7, 2, 5], flow shop problems with limited storage and resource constraints have not been thoroughly studied before. Also, our problem may be thought of as a special packet routing problem on a network of three nodes with a bounded queue at each node [14] no relevant results exist for our special ....

....general problem may be thought of as 3 machine flow shop scheduling with limited storage and additional resource constraints. This problem is clearly NP hard. In general, scheduling problems with additional resource constraints do not have small constant factor approximations (see for example, [2, 7]) We propose the simple algorithmic approach of solving the P = D = T = 1 version of the problem and simulating the order of job execution in parallel for the general case. Although this approach of simulating the sequential schedule with multiple machines is a provably good approximation in some ....

Garofalakis, M. N., and Ioannidis, Y. E. Parallel query scheduling and optimization with time- and space-shared resources. In Proceedings of VLDB '97 (Athens, Greece, Aug. 1997), pp. 296--305.

....operators are malleable in the sense that they can operate in a range of memory allocations between their minimum and maximum requirements. The task of determining the best query plan which takes into account this malleability of operators is generally considered a hard optimization problem [4]. In fact, this problem is rather similar to that of query optimization for parallel database systems. 5] mentions that one can either employ a one phase (optimization combined with parallelization) or a two phase (optimization followed by parallelization) strategy. The second method is ....

....parallel database systems. The second area of related research concentrates on the optimal scheduling of operators and queries in a parallel system. 7] and [12] both propose strategies for determining the degree of parallelism and choosing the nodes on which the operators of the query should run. [4] also provides scheduling algorithms based on multi dimensional bin packing that take into account both time shared (CPU, disk, network) and space shared (memory) resources. Despite the immense amount of work done in this field, there have been very few studies dealing with the execution of joins ....

[Article contains additional citation context not shown here]

M.N. Garofalakis and Y.E. Ioannidis. "Parallel Query Scheduling and Optimization with Time- and SpaceShared Resources". Proc. of VLDB Conf. 1997.

....time algorithm can achieve a better approximation ratio. Besides having theoretical importance, these problems have several applications such as load balancing, cutting stock, and resource allocation, to name a few. One of our motivations for studying these problems comes from recent interest [14, 15, 16] in multidimensional resource scheduling problems in parallel query optimization. A favored architecture for parallel databases is the so called shared nothing environment [7] where the parallel system consists of a set of indepen1 2 dent processing units each of which has a set of timeshareable ....

....by a single aggregate work measure. This simplification is done typically to reduce the complexity of the scheduling problem. For large task systems that are typically encountered in database applications, however, ignoring the multi dimensionality could lead to bad performance. The work in [13, 14, 15, 16] demonstrates the practical effectiveness of the multi dimensional approach. One of the basic resource scheduling problem that is considered in the above papers is the problem of scheduling d dimensional vectors (tasks) on d dimensional bins (machines) to minimize the maximum load on any ....

[Article contains additional citation context not shown here]

Minos N. Garofalakis and Yannis E. Ioannidis. Parallel query scheduling and optimization with time-and space-shared resources. In Proceedings of the 23rd VLDB Conference, pages 296--305, 1997.

No context found.

M. Garofalakis and Y. Ioannidis. Parallel query scheduling and optimization with time- and space-shared resources. In VLDB, 1997.

....requirements of the presentation over time with their enclosing Minimum Bounding Rectangle (MBR) Although this simplification significantly reduces the complexity of the relevant scheduling problems, it suffers from two major deficiencies. ffl The volume (i.e. resource time product [CM96, GI97] in the enclosing MBR can be significantly larger than the actual requirements of the composite object. This can result in wasting large fractions of precious server resources, especially for relatively sparse composite objects. ffl The MBR simplification hides the timing structure of ....

....sequence: l i 1 ; r i 1 ) l i k i ; r i k i ) where k i l(C i ) In fact, k i 2 Delta n i Gamma 1, where n i is the number of component CM streams in C i . We define the volume (V) of a composite object C i as the total resource time product over the duration of C i [CM96, GI97] More formally, V (C i ) P k i j=1 l i j r i j . The density (d) of a composite object C i is defined as the ratio of the object s volume to the volume of its MBR, i.e. d(C i ) V (C i ) l(C i ) Deltar max (C i ) 2.2 Using Memory to Change Object Sequences: Stream Sliding Although ....

[Article contains additional citation context not shown here]

Minos N. Garofalakis and Yannis E. Ioannidis. "Parallel Query Scheduling and Optimization with Time- and Space-Shared Resources". In Proceedings of the 23rd International Conference on Very Large Data Bases, pages 296--305, Athens, Greece, August 1997.

....parameters of parallel execution. Note to referees: Parts of this paper have appeared in the Proceedings of the 1996 ACM SIGMOD International Conference on Management of Data (SIGMOD 96) GI96] and in the Proceedings of the 23rd International Conference on Very Large Data Bases (VLDB 97) GI97] Besides the more extensive coverage, this paper extends these earlier conference publications with significant new material, including: a) experimental results from an implementation of our scheduling algorithms on top of a detailed simulation model of a parallel DBMS (Section 7) and (b) ....

Minos N. Garofalakis and Yannis E. Ioannidis. "Parallel Query Scheduling and Optimization with Timeand Space-Shared Resources". In Proceedings of the 23rd International Conference on Very Large Data Bases, pages 296--305, Athens, Greece, August 1997.

....resource requirements of the presentation over time with their enclosing Minimum Bounding Rectangle (MBR) Although this simplification significantly reduces the complexity of the relevant scheduling problems, it suffers from two major deficiencies. ffl The volume (i.e. resource time product [16]) in the enclosing MBR can be significantly larger than the actual requirements of the composite object. This can result in wasting large fractions of precious server resources, especially for relatively sparse composite objects. ffl The MBR simplification hides the timing structure of ....

....by the sequence: l i 1 ; r i 1 ) l i k i ; r i k i ) where k i l(C i ) In fact, k i 2 Delta n i Gamma 1, where n i is the number of component CM streams in C i . We define the volume (V) of a composite object C i as the total resource time product over the duration of C i [16]. More formally, V (C i ) P k i j=1 l i j r i j . The density (d) of a composite object C i is defined as the ratio of the object s volume to the volume of its MBR, i.e. d(C i ) V (C i ) l(C i ) Deltar max (C i ) 2.2 Using Memory to Change Object Sequences: Stream Sliding Although ....

[Article contains additional citation context not shown here]

M.N. Garofalakis and Y.E. Ioannidis. "Parallel Query Scheduling and Optimization with Time- and Space-Shared Resources". In Proc. of the 23rd Intl. Conf. on Very Large Data Bases, August 1997.

....implications of our results for the open problem of designing efficient cost models for parallel query optimization [7] 1 Due to space constraints, we do not discuss the details of earlier work. For an extensive bibliography, the interested reader is referred to the full version of the paper [12]. 2 Problem Formulation 2.1 Definitions We consider hierarchical parallel systems [2] with identical multiprogrammed resource sites connectedby an interconnection network. Each site is a collection of d TS resources (e.g. CPU(s) disk(s) and network interface(s) or communication ....

....on P sites then T par (OPT;P ) LB(S; P ) where LB(S;P ) max ae T max (S) l(S W ) P ; l(S TV ) P oe : As with all theoretical results presented here, Theorem 4.1 is stated without proof due to space constraints. The details can be found in the full version of this paper [12]. Compared to our earlier results [11] the lower bound in Theorem 4.1 introduces a third term containing l(S TV ) i.e. the total volume of the parallel execution. We will see that this new parameter plays an important role in our analytical and experimental results. The basic idea of our ....

[Article contains additional citation context not shown here]

M. N. Garofalakis and Y. E. Ioannidis. "Parallel Query Scheduling and Optimization with Time- and Space-Shared Resources". Unpublished manuscript, June 1997.

No context found.

M. N. Garofalakis and Y. E. Ioannidis. Parallel query scheduling and optimization with time- and space-shared resources. In VLDB, pages 296--305, 1997.

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