Efficient implementations of search trees on parallel distributed memory architectures

Colbrook, Adrian; Smythe, C.

doi:10.1049/ip-e.1990.0049

Cited by 10 publications

(7 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…16 First, we show that the multi-version memory algorithm does not allow find x operations to fail (which, as mentioned earlier, would cause operations to halt). Second, we show that if the dictionary data structure in the coherent shared memory algorithm Therefore, the read step for n iϩ1 inserted into S reads version B.…”

Section: A3 Liveness Propertiesmentioning

confidence: 62%

“…This old version may redirect the operation back to Z2, and this cycle can be repeated. This is why our transformed find x operation issues a readcurrent to any node already visited (lines [15][16][17]. In Appendix A, we prove that this protocol avoids cycles.…”

Section: Performance Improvementsmentioning

confidence: 95%

“…This includes proposing new algorithms (e.g., [4,22,40,10,43,48,38,16,15,30,31]) and comparing the performance of existing algorithms (e.g., [39,33,53,51]). Most of these studies discount the effects of memory replication and caching, or ignore the issue entirely.…”

Section: Appendix A: Transformation Proof Of Correctnessmentioning

confidence: 99%

“…As a result, many researchers (e.g., [4,22,40,10,43, 48,38,16,15,30,31,32]) have developed B-tree algorithms that support concurrent operations.2 Concurrent B-tree algorithms require concurrency control: operations that update the state of the tree must not be allowed to interfere with other operations.There are two general strategies for maintaining concurrency control in a B-tree: lock-coupling and link-method algorithms. Both require each node in the tree to be associated with a read/write lock [4].…”

mentioning

confidence: 99%

See 3 more Smart Citations

Scalable Concurrent B-Trees Using Multi-Version Memory

Wang

Weihl²

1996

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Section: A3 Liveness Propertiesmentioning

confidence: 62%

Section: Performance Improvementsmentioning

confidence: 95%

Section: Appendix A: Transformation Proof Of Correctnessmentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Scalable Concurrent B-Trees Using Multi-Version Memory

Wang

Weihl²

1996

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

“…Colbrook and Smythe devised a balanced search tree called 2 P −2 − 2 P [15] for distributed-memory parallel machines, and Colbrook et al [16] introduced a variation of a distributed B-tree for processes connected in a ring topology. Tu et al [17] used the same parallel octree and regular grid decomposition for parallel volume rendering as was used to compute a seismic finite-element simulation.…”

Section: Related Workmentioning

confidence: 99%

Efficient Delaunay Tessellation through K-D Tree Decomposition

Morozov¹,

Peterka²

2016

SC16: International Conference for High Performance Computing, Networking, Storage and Analysis

View full text Add to dashboard Cite

Delaunay tessellations are fundamental data structures in computational geometry. They are important in data analysis, where they can represent the geometry of a point set or approximate its density. The algorithms for computing these tessellations at scale perform poorly when the input data is unbalanced. We investigate the use of k-d trees to evenly distribute points among processes and compare two strategies for picking split points between domain regions. Because resulting point distributions no longer satisfy the assumptions of existing parallel Delaunay algorithms, we develop a new parallel algorithm that adapts to its input and prove its correctness. We evaluate the new algorithm using two late-stage cosmology datasets. The new running times are up to 50 times faster using k-d tree compared with regular grid decomposition. Moreover, in the unbalanced data sets, decomposing the domain into a k-d tree is up to five times faster than decomposing it into a regular grid.

show abstract

Algorithms for search trees on message passing architectures

Colbrook

Brewer²,

Dellarocas³

et al. 1996

IEEE Trans. Parallel Distrib. Syst.

Self Cite

View full text Add to dashboard Cite

In this paper we describe a new algorithm for maintaining a balanced search tree on a message-passing "MD architecture; the algorithm is particularly well suited for implementation on a small number of processors. We introduce a (28', 2 B) search tree that uses a linear array of t(log n) processors to store n entries. Update operations use a bottom-up node-splitting scheme, which performs better than top-down search tree algorithms. Additionally, for a given cost ratio of computation to coununication the value of B may be varied to m-anize performance. Implementations on a parallel-architecture simulator are described. 20 DISTRIBUTION/AVAiLABILITY OF ABSTRACT 21. ABSTRACT SECURITY CLASSIFICATION (3 UNCLASSIFiEDIUNLIMITED I2 SAME AS RPT. 0 OTIC USERS Unclassified 22a NAME OF RESPONSIBLE INDIVIDUAL 22b. TELEPHONE (Include Area Code.i 22c. OFFICE SYMBOL

show abstract

Efficient implementations of search trees on parallel distributed memory architectures

Cited by 10 publications

References 2 publications

Scalable Concurrent B-Trees Using Multi-Version Memory

Scalable Concurrent B-Trees Using Multi-Version Memory

Efficient Delaunay Tessellation through K-D Tree Decomposition

Algorithms for search trees on message passing architectures

Contact Info

Product

Resources

About