MUMPS: A General Purpose Distributed Memory Sparse Solver

Amestoy, Patrick; Duff, Iain S.; L’Excellent, Jean-Yves; Koster, Jacko

doi:10.1007/3-540-70734-4_16

Cited by 144 publications

(108 citation statements)

References 7 publications

Supporting

Mentioning

108

Contrasting

Order By: Relevance

“…The computation of the stationary distribution has been solved using the linear algebra library MUMPS ( [1]). Many packet payload sizes were considered as well as two data rates, 11 Mb/s (corresponding to the maximum data rate of IEEE 802.11b) and 2 Mb/s (corresponding to the maximum data rate of the original IEEE 802.11).…”

Section: Resultsmentioning

confidence: 99%

Study of the impact of asymmetry and carrier sense mechanism in IEEE 802.11 multi-hops networks through a basic case

Chaudet

Lassous

Thierry

et al. 2004

Proceedings of the 1st ACM International Workshop on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks

View full text Add to dashboard Cite

This article presents a theoretical modeling of a small but basic scenario that shows a great inequity in medium access between nodes using the IEEE 802.11 DCF mode. In this configuration, two terminals evolve independently and are almost never synchronized, which results in a serious performance issue for a third emitter in between. The results are compared to simulation results and a discussion on the different performance loss causes follows.

show abstract

Section: Resultsmentioning

confidence: 99%

Study of the impact of asymmetry and carrier sense mechanism in IEEE 802.11 multi-hops networks through a basic case

Chaudet

Lassous

Thierry

et al. 2004

Proceedings of the 1st ACM International Workshop on Performance Evaluation of Wireless Ad Hoc, Sensor, and Ubiquitous Networks

View full text Add to dashboard Cite

show abstract

“…These interfaces allow Epetra matrices and vectors to be used with each third-party solver. At this time, we provide support for SuperLU (serial), SuperLUDist [Li and Demmel 2003], Kundert's Sparse solver (from Spice [Quarles et al 2003]),DSCPack [Raghavan 2003], UMFPack [Davis 2003] and MUMPS [Amestoy et al 2003]. …”

Section: Amesos: Object-oriented Interface To Direct Solversmentioning

confidence: 99%

An overview of the Trilinos project

Heroux

Bartlett

Howle

et al. 2005

ACM Trans. Math. Softw.

952

671

View full text Add to dashboard Cite

The Trilinos Project is an effort to facilitate the design, development, integration and ongoing support of mathematical software libraries within an object-oriented framework for the solution of large-scale, complex multi-physics engineering and scientific problems. Trilinos addresses two fundamental issues of developing software for these problems: (i) Providing a streamlined process and set of tools for development of new algorithmic implementations and (ii) promoting interoperability of independently developed software. Trilinos uses a two-level software structure designed around collections of packages. A Trilinos package is an integral unit usually developed by a small team of experts in a particular algorithms area such as algebraic preconditioners, nonlinear solvers, etc. Packages exist underneath the Trilinos top level, which provides a common look-and-feel, including configuration, documentation, licensing, and bug-tracking.Here we present the overall Trilinos design, describing our use of abstract interfaces and default concrete implementations. We discuss the services that Trilinos provides to a prospective package and how these services are used by various packages. We also illustrate how packages can be combined to rapidly develop new algorithms. Finally, we discuss how Trilinos facilitates highquality software engineering practices that are increasingly required from simulation software.

show abstract

“…This may be a subproblem within a larger parallel context. Some parallel sparse direct solvers (e.g., SuperLU-MT [5], [6] or Pardiso [7]) have shown good performance in shared-memory environments, while distributed-memory solvers (for example MUMPS [8], [9]) have limited scalability. Pastix [10] is an interesting sparse direct solver because it uses hybrid parallel programming with both MPI and threads.…”

Section: B Previous Workmentioning

confidence: 99%

ShyLU: A Hybrid-Hybrid Solver for Multicore Platforms

Rajamanickam

Boman

Heroux

2012

2012 IEEE 26th International Parallel and Distributed Processing Symposium

View full text Add to dashboard Cite

Abstract-With the ubiquity of multicore processors, it is crucial that solvers adapt to the hierarchical structure of modern architectures. We present ShyLU, a "hybrid-hybrid" solver for general sparse linear systems that is hybrid in two ways: First, it combines direct and iterative methods. The iterative part is based on approximate Schur complements where we compute the approximate Schur complement using a value-based dropping strategy or structure-based probing strategy.Second, the solver uses two levels of parallelism via hybrid programming (MPI+threads). ShyLU is useful both in sharedmemory environments and on large parallel computers with distributed memory. In the latter case, it should be used as a subdomain solver. We argue that with the increasing complexity of compute nodes, it is important to exploit multiple levels of parallelism even within a single compute node.We show the robustness of ShyLU against other algebraic preconditioners. ShyLU scales well up to 384 cores for a given problem size. We also study the MPI-only performance of ShyLU against a hybrid implementation and conclude that on present multicore nodes MPI-only implementation is better. However, for future multicore machines (96 or more cores) hybrid/ hierarchical algorithms and implementations are important for sustained performance.

show abstract

MUMPS: A General Purpose Distributed Memory Sparse Solver

Cited by 144 publications

References 7 publications

Study of the impact of asymmetry and carrier sense mechanism in IEEE 802.11 multi-hops networks through a basic case

Study of the impact of asymmetry and carrier sense mechanism in IEEE 802.11 multi-hops networks through a basic case

An overview of the Trilinos project

ShyLU: A Hybrid-Hybrid Solver for Multicore Platforms

Contact Info

Product

Resources

About