Chang, C., Kurc, T., Sussman, A., Catalyurek, U., and Saltz, J.: A hypergraphbased workload partitioning strategy for parallel data aggregation. In Proc. of 11th SIAM Conf. Parallel Processing for Scientific Computing. SIAM, (2001)  Home/Search   Document Details and Download   Summary   Related Articles   Check

....operator involves the selection of a sub dataset from a data partition followed by a transformation that produces the output data from the selected sub dataset. On the other hand, we do assume the # REDUCE operator to be both commutative and associative, which is generally the case in practice [4, 13, 18, 22]. As will be shown in the later sections, requiring the #REDUCE operator to be commutative and associative allows us to perform parallel reduction with limited synchronization. 3.2 Adding Quality Control to DAC The specification in Figure 2 assumes that there is no cluster node failures and the ....

....achieve the efficiency offered by the event driven concurrency management. MPI [22] also supports data reduction operations and several previous works have studied tree based MPI reductions [10, 11, 12, 23] In addition, parallel data aggregation has been studied in scientific computation research [4], where multiple datasets are mapped to multiple processors and all processors perform aggregations on local sub datasets in parallel. Our DAC primitive targets service programming domain and differs from the MPI reduce primitive significantly. MPI reduce does not tolerate node failures, nor does ....

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraph-based workload partitioning strategy for parallel data aggregation. In SIAM PPSC, Portsmouth, Virginia, Mar. 2001.

No context found.

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraph-based workload partitioning strategy for parallel data aggregation. In Proceedings of the Eleventh SIAM Conference on Parallel Processing for Scientific Computing. SIAM, Mar. 2001.

No context found.

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraph-based workload partitioning strategy for parallel data aggregation. In Proceedings of the Eleventh SIAM Conference on Parallel Processing for Scientific Computing. SIAM, Mar. 2001.

....the speedup for five server processors is 3.6 compared to a one processor server. 12 3.3 Query Processing Strategies Workload partitioning and tiling have significant effects on the performance of an application implemented using the ADR framework. We have evaluated several potential strategies [22,23,43] that use different workload partitioning and tiling schemes. To simplify the presentation, we assume that the target range query involves only one input and one output dataset. Both the input and output datasets are assumed to be already partitioned into data chunks and declustered across the ....

....input chunks that map to the same output chunk must be forwarded to the processor that owns the output chunk. Since a projection function may map an input chunk to multiple output chunks, an input chunk may be forwarded to multiple processors. 3.3. 4 A Hypergraph based Strategy In this strategy [22], workload partitioning is formulated as a hypergraph partitioning problem. A hypergraph is a graph where each hyperedge (also called a net) can connect more than two vertices in the graph. We first describe the tiling algorithm, and then the workload partitioning algorithm. Tiling: The memory ....

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraphbased workload partitioning strategy for parallel data aggregation. In Proceedings of the Eleventh SIAM Conference on Parallel Processing for Scientific Computing. SIAM, Mar. 2001.

....the use of SMP clusters to improve response times and overall system performance. In particular, we look at the effective use of aggregate processing power and I O bandwidth for executing single and multiple queries efficiently. Unlike previous work on query execution in parallel systems [5, 6, 10, 12], our system design combines parallel execution of queries with data caching and multi threaded execution so that multiple queries can execute concurrently on multiple processors on an SMP node and also reuse cached results to improve performance and lower interprocess communication. Moreover, a ....

....results. This scheme can effectively result in a partitioningof the accumulator data structure across the nodes. Both schemes are 2 We have used the MPICH implementation of MPI, which is not threadsafe. shown schematically in Figure 1(b) Other strategies are also possible. Our earlier work [6, 10] and the work of Shatdal and Naughton [12] have shown that other strategies, such as distributed accumulator, may outperform the replicated accumulator strategies, depending on machine configuration (e.g. number of nodes) and application characteristics. The previous work evaluated various ....

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraph-based workload partitioning strategy for parallel data aggregation. In Proceedings of the Eleventh SIAM Conferenceon Parallel Processing for Scientific Computing. SIAM, Mar. 2001.

No context found.

Chang, C., Kurc, T., Sussman, A., Catalyurek, U., and Saltz, J.: A hypergraphbased workload partitioning strategy for parallel data aggregation. In Proc. of 11th SIAM Conf. Parallel Processing for Scientific Computing. SIAM, (2001)

No context found.

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, J. Saltz, A hypergraph-based workload partitioning strategy for parallel data aggregation, in: Proc. of 11th SIAM Conf. Parallel Processing for Scientific Computing, SIAM, 2001.

No context found.

C. Chang, T. Kurc, A. Sussman, U. Catalyurek, and J. Saltz. A hypergraph-based workload partitioning strategy for parallel data aggregation. In SIAM PPSC, Portsmouth, Virginia, Mar. 2001.

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