Parallel Partition Revisited

Frias, Leonor; Petit, Jordi

doi:10.1007/978-3-540-68552-4_11

Cited by 7 publications

(11 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…On the one hand, partition compares each element exactly once against the pivot. That is guaranteed precisely by the parallel partitioning algorithms in [6] . On the other hand, the implicit structure defined by the recursive calls corresponds to a BST.…”

Section: Putting All Togethermentioning

confidence: 94%

“…A summary of existing parallel partitioning algorithms was presented in [6]. These include a simple algorithm by Francis and Pannan [5] (STRIDED), a fetch-and-add algorithm by Tsigas and Zhang [13] and a variation of the former avaliable in the MCSTL library [10] (F&A).…”

Section: Parallel Partition For Atomic Keysmentioning

confidence: 99%

“…Though very different in nature, these algorithms can be divided into three main phases: a) A sequential setup of each processor's work, b) a parallel main phase in which most of the partitioning is done, and c) a cleanup phase, which is usually sequential. In [6] it was noted that all the above algorithms disregard part of the work done in the main parallel phase when cleaning up. In order to overcome this drawback, an alternative parallel cleanup phase that uses the whole comparison information of the parallel phase was proposed.…”

Section: Parallel Partition For Atomic Keysmentioning

confidence: 99%

“…The parallel partitioning algorithms in [6] are provided according to the specification of the partition algorithm of the Standard Template Library (STL) of the C ++ programming language [8]. The input parameters of partition and their based algorithm, sort and nth element, are a sequence and a comparator.…”

Section: Outer Changesmentioning

confidence: 99%

“…In parallelizing quicksort and quickselect, a key problem is parallelizing the partition. Recently, in [6] a summary of existing parallel partitioning algorithms and a further modification to obtain the minimal number of comparisons (i.e. n) has been presented.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Combining digital access and parallel partition for quicksort and quickselect

Frias

Petit

2009

2009 ICSE Workshop on Multicore Software Engineering

Self Cite

View full text Add to dashboard Cite

In this paper, we combine digital access to strings with parallel partition to enhance parallel quicksort and quickselect implementations. Previously, digital access had only been combined with sequential comparison-based algorithms and data structures. We present broadly our approach, not only algorithmically but from the design and implementation point of view. Finally, we give some experimental results that indicate its effectiveness in practice.

show abstract

Section: Putting All Togethermentioning

confidence: 94%

Section: Parallel Partition For Atomic Keysmentioning

confidence: 99%

Section: Parallel Partition For Atomic Keysmentioning

confidence: 99%

Section: Outer Changesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Combining digital access and parallel partition for quicksort and quickselect

Frias

Petit

2009

2009 ICSE Workshop on Multicore Software Engineering

Self Cite

View full text Add to dashboard Cite

show abstract

An OpenMP Parallelization of the K-means Algorithm Accelerated Using KD-trees

Kwedlo

Łubowicz

2020

Parallel Processing and Applied Mathematics

View full text Add to dashboard Cite

Accelerated K-Means Algorithms for Low-Dimensional Data on Parallel Shared-Memory Systems

Kwedlo

Łubowicz

2021

IEEE Access

View full text Add to dashboard Cite

This paper considers the problem of exact accelerated algorithms for the K-means clustering of low-dimensional data on modern multi-core systems. A version of the filtering algorithm parallelized using the OpenMP (Open Multi-Processing) standard is proposed. The algorithm employs a kd-tree structure to skip some unnecessary calculations between cluster centroids and feature vectors. In our approach, both the kd-tree construction and the iterations of the K-means are parallelized using the OpenMP tasking mechanism. A new task is created for a recursive call performed during kd-tree construction and traversal. The tasks are executed in parallel by the cores of a shared-memory system. In computational experiments, we evaluated the parallel efficiency of our approach and compared its performance to the parallel Lloyd's method, a GPU (Graphics Processing Unit) formulation of the K-means algorithm, and two parallel triangle inequality-based algorithms intended for low-dimensional data. The evaluation was performed on six synthetic datasets from two distributions and seven real-life datasets. The experiments, executed on a 24core system, indicated that our version of the filtering algorithm had satisfactory or high parallel efficiency. Its runtime was much shorter than those of competing algorithms. However, the advantage of the parallel filtering algorithm decreased rapidly as the dimension of data increased.INDEX TERMS acceleration of K-means, K-means clustering, kd-trees, OpenMP tasks, parallelization

show abstract

Parallel Partition Revisited

Cited by 7 publications

References 6 publications

Combining digital access and parallel partition for quicksort and quickselect

Combining digital access and parallel partition for quicksort and quickselect

An OpenMP Parallelization of the K-means Algorithm Accelerated Using KD-trees

Accelerated K-Means Algorithms for Low-Dimensional Data on Parallel Shared-Memory Systems

Contact Info

Product

Resources

About