Graph Partitioning for Dynamic, Adaptive and Multi-Phase Scientific Simulations

Schloegel, Kirk; Karypis, George; Kumar, Vijayendra

doi:10.1142/9781860949630_0004

Cited by 89 publications

(135 citation statements)

References 0 publications

Supporting

Mentioning

135

Contrasting

Order By: Relevance

“…For example, solving the graph partitioning problem can help to balance load and minimize communication in scientific simulations [150,35,69], can speed up Dijkstra's Algorithm [108,120], and in general is an useful technique in the route planning area [111,100,48]; it supports VLSI design [7,8], and also can preserve sparsity in Gaussian elimination on sparse symmetric positive definite matrices [74]. Probably the best known application of graph partitioning is the numerical solution of partial differential equations on a highly parallel computer.…”

Section: Motivationmentioning

confidence: 99%

“…Indeed, high quality solutions are also favorable in applications where the graph needs to be partitioned only once so that the partition can be used over and over again and the running time of the graph partitioning algorithms is only a minor issue [108,120,111,100,48,69]. Thirdly, high quality solutions are even important in areas in which the running time overhead is paramount [157], such as finite element computations [150] or the direct solution of sparse linear systems [74]. Here, high quality graph partitions can be useful for benchmarking purposes.…”

Section: Motivationmentioning

confidence: 99%

“…There has been a huge amount of research on graph partitioning so that we refer the reader to [70,150,23] for most of the material. Hence, in this chapter we try to focus on selected related work and recent coordinate free techniques that have not yet been covered by these papers.…”

Section: Related Workmentioning

confidence: 99%

See 2 more Smart Citations

High quality graph partitioning

Sanders¹,

Schulz²

2013

Graph Partitioning and Graph Clustering

View full text Add to dashboard Cite

Section: Motivationmentioning

confidence: 99%

Section: Motivationmentioning

confidence: 99%

See 1 more Smart Citation

High quality graph partitioning

Sanders¹,

Schulz²

2013

Graph Partitioning and Graph Clustering

View full text Add to dashboard Cite

“…Parallel formulations of multilevel graph partitioning schemes have also been proposed in the literature, although their development is quite challenging. Moreover, parallel versions of the three well known partition packages Jostle [48], Metis [40] and Scotch [36], based on the multilevel paradigm, have also been developed. Perhaps the fastest available parallel code is the parallel version of Metis (parMatis).…”

Section: Classical Approaches For the Graph Partitioning Problemmentioning

confidence: 99%

Hybrid Metaheuristics for the Graph Partitioning Problem

Benlic

Hao

2013

Hybrid Metaheuristics

View full text Add to dashboard Cite

The Graph Partitioning Problem (GPP) is one of the most studied NPcomplete problems notable for its broad spectrum of applicability such as in VLSI design, data mining, image segmentation, etc. Due to its high computational complexity, a large number of approximate approaches have been reported in the literature. Hybrid algorithms that are based on adaptations of popular metaheuristic techniques have shown to provide outstanding performance in terms of partition quality. In particular, it is the hybrids between well-known metaheuristics and multilevel strategies that report partitions of the minimal cut-size value. However, metaheuristic hybrids generally require more computing time than those based on greedy heuristics which can generate partitions of acceptable quality in a matter of seconds even for very large graphs. This chapter is dedicated to a review on some representative hybrid metaheuristic approaches including genetic local search, basic multilevel search and recent development on hybrid multilevel search.

show abstract

“…The current implementation of SDM requires that the partition vector fit in memory; an out-of-core partition vector is not currently supported. For the application illustrated in Figure 6, we used the partitioning vector generated from the graph-partitioning tool Metis [20,34].…”

Section: Implementation Detailsmentioning

confidence: 99%

High-performance scientific data management system

No¹,

Thakur

Choudhary

2003

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

Many scientific applications have large I/O requirements, in terms of both the size of data and the number of files or data sets. Management, storage, efficient access, and analysis of this data present an extremely challenging task. Traditionally, two different solutions have been used for this task: file I/O or databases. File I/O can provide high performance but is tedious to use with large numbers of files and large and complex data sets. Databases can be convenient, flexible, and powerful but do not perform and scale well for parallel supercomputing applications. We have developed a software system, called Scientific Data Manager (SDM), that combines the good features of both file I/O and databases. SDM provides a high-level API to the user and, internally, uses a parallel file system to store real data (using various I/O optimizations available in MPI-IO) and a database to store application-related metadata. In order to support I/O in irregular applications, SDM makes extensive use of MPI-IO's noncontiguous collective I/O functions. Moreover, SDM uses the concept of a history file to optimize the cost of the index distribution using the metadata stored in database. We describe the design and implementation of SDM and present performance results with two regular applications, ASTRO3D and an Euler solver, and with two irregular applications, a CFD code called FUN3D and a Rayleigh-Taylor instability code.

show abstract

Graph Partitioning for Dynamic, Adaptive and Multi-Phase Scientific Simulations

Cited by 89 publications

References 0 publications

High quality graph partitioning

High quality graph partitioning

Hybrid Metaheuristics for the Graph Partitioning Problem

High-performance scientific data management system

Contact Info

Product

Resources

About