M.J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29, 2 (1989), 239--250. 24  Home/Search   Document Not in Database   Summary   Related Articles   Check

....by overpartitioning [21] The price paid by these single step algorithms is an irregular communication scheme and difficulty with load balancing. The other group of sorting algorithms is the multi step algorithms, which include bitonic sort [8] column sort [20] rotate sort [23] hyperquicksort [26], flashsort [27] B flashsort [18] smoothsort [25] and Tridgell and Brent s sort [30] Generally speaking, these algorithms accept multiple rounds of communication in return for better load balancing and, in some cases, regular communication. In this paper, we present a novel variation on the ....

M.J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29, 2 (1989), 239--250. 24

....by overpartitioning [21] The price paid by these single step algorithms is an irregular communication scheme and difficulty with load balancing. The other group of sorting algorithms is the multi step algorithms, which include bitonic sort [8] column sort [20] rotate sort [23] hyperquicksort [26], flashsort [27] B flashsort [18] smoothsort [25] and Tridgell and Brent s sort [30] Generally speaking, these algorithms accept multiple rounds of communication in return for better load balancing and, in some cases, regular communication. In this paper, we present a novel variation on the ....

M. J. Quinn, Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge, BIT 29, No. 2 (1989), 239#250.

....sorting programs based on the PSRS (Parallel Sampling by Regular Sampling) technique and under the framework of 4 See http: www.umiacs.umd.edu research EXPAR 5 BSP. To our knowledge, little work has been performed about sorting on non homogeneous network. If we try to reuse the main results [10, 3, 2, 4] about sorting on homogeneous clusters as a general strategy, the main diculty is to show that it will lead to good properties for load balancing, for execution time and also that the underlying algorithm is suitable for a concrete implementation. We are not yet convinced that other technique than ....

M.J. Quinn, \Analysis and benchmarking of two parallel sorting algorithms: Hyperquicksort and quickmerge", BIT, vol. 29, no. 2, pp. 239-250, 1989. 6

....by overpartitioning [21] The price paid by these single step algorithms is an irregular communication scheme and difficulty with load balancing. The other group of sorting algorithms is the multi step algorithms, which include bitonic sort [8] column sort [20] rotate sort [23] hyperquicksort [26], flashsort [27] B flashsort [18] smoothsort [25] and Tridgell and Brent s sort [30] Generally speaking, these algorithms accept multiple rounds of communication in return for better load balancing and, in some cases, regular communication. In this paper, we present a novel variation on the ....

M.J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29, 2 (1989), 239--250.

....dimension of the subcube) This is applied recursively within each subcube. It is very important to pick a pivot that closely balances the two halves otherwise the load on the processors can become extremely imbalanced. Hyperquicksort has been implemented on the NCUBE 10 [39] and the NCUBE 7 [29]. On the NCUBE it was shown to be about twice as fast as bitonic sort, but this was based on randomly generated keys (which would be expected to be a good distribution for the sort) Another variation of quicksort allocates a fair number of processors to each recursive call so that picking bad ....

M. J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29(2):239--250, 1989.

....send must have a matching receive. The primitives transfer data blocks via DMA over the network to the target processor. There is no cache coherence. The algorithm comparison comes from Quinn s paper Analysis and Benchmarking of Two Parallel Sorting Algorithms: Hyperquicksort and Quickmerge [Qui89] Quinn compares two sorting algorithms on a 64 processor nCUBE 7. Both algorithms mix local sorting with communication; they differ in their strategies for dividing the values among the processors. In general, quickmerge requires fewer but larger messages than hyperquicksort. Figure 3 graphs ....

M. J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: Hyperquicksort and quickmerge. BIT, 29(2):239--250, 1989.

....of communication rounds, provided p 3 n. Indeed, there are a host of published algorithms for achieving such a result when the ratio of input size to number of processors is as large as this. For example, a randomized strategy, called sample sort, achieves this result with high probability [8, 19, 20, 23, 24, 32, 40, 42], as do deterministic strategies based upon regular sampling [18, 36, 43, 46] These methods based upon sampling do not seem to scale nicely for smaller n=p ratios, however. If columnsort is implemented in a recursive fashion, then it yields an EREW BSP algorithm that uses TC = O( log n= ....

M. J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29(2):239--250, 1989.

....by overpartitioning [23] The price paid by these single step algorithms is an irregular communication scheme and difficulty with load balancing. The other group of sorting algorithms is the multi step algorithms, which include bitonic sort [9] column sort [22] rotate sort [25] hyperquicksort [28], flashsort [29] B flashsort [20] smoothsort [27] and Tridgell and Brent s sort [32] Generally speaking, these algorithms accept multiple rounds of communication in return for better load balancing and, in some cases, regular communication. In this paper, we present a novel variation on the ....

M.J. Quinn. Analysis and Benchmarking of Two Parallel Sorting Algorithms: Hyperquicksort and Quickmerge. BIT, 29:239--250, 1989.

....by overpartitioning [24] The price paid by these single step algorithms is an irregular communication scheme and difficulty with load balancing. The other group of sorting algorithms is the multi step algorithms, which include bitonic sort [9] column sort [23] rotate sort [26] hyperquicksort [29], flashsort [30] B flashsort [19] smoothsort [28] and Tridgell and Brent s sort [33] Generally speaking, these algorithms accept multiple rounds of communication in return for better load balancing and, in some cases, regular communication. In this paper, we present a novel variation on the ....

M.J. Quinn. Analysis and Benchmarking of Two Parallel Sorting Algorithms: Hyperquicksort and Quickmerge. BIT, 29:239--250, 1989.

....[4] and parallel algorithms for sorting have been studied intensively ever since Batcher [5] proposed the bitonic sorting network. Although the theoretical community has devoted considerable effort to the design of parallel sorting algorithms [2, 3, 18, 27, 10, 26, 16, 23] empirical studies [12, 11, 7, 34, 25, 29, 35] started appearing only in the late 1980 s when multiprocessors became widely available. In a typical parallel sorting algorithm, each processor contains a portion of the list of n elements to be sorted. In a merge based approach, each processor sorts the portion it holds initially by some ....

....and Engineering Research Council of Canada and from the Information Technology Research Centre of Ontario. 1 A single step algorithm will merge the portions in all the processors or partition the sublists and assign to all the processors in one step. Single step Multi step merge quickmerge [25] Batcher s bitonic sort [5] based PSRS [30, 21] parallel merge sort [9, 10] smoothsort [24] Nassimi and Sahni s sort [22] column sort [18] Tridgell and Brent s sort [32] snakesort [6] quicksort parallel quicksort [7, 20, 13] flashsort [27] based parallel sample sort [15, 8] B flashsort [14] ....

[Article contains additional citation context not shown here]

M. J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29(2):239--250, 1989.

No context found.

M.J. Quinn, Analysis and Benchmarking of Two Parallel Sorting Algorithms: Hypersort and Quickmerge, BIT, 29(2):239-250, 1989.

No context found.

M. J. Quinn. Analysis and benchmarking of two parallel sorting algorithms: hyperquicksort and quickmerge. BIT, 29(2):239--250, 1989.

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