K. Burrage and B. Pohl, Implementing an ODE code on distributed memory computers, Computers Math. Applic., 28, (1994), 235--252. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
Numerical experiments with some explicit pseudo two-step.. - Cong, Podhaisky, Weiner (1998) (Correct)
....with fast data access requiring less communication times which is suited to the features of the EPTRK methods. In addition, there are the advantages of compilers which attempt to parallelize codes automatically by reordering loops and sophisticated scientific libraries (cf. e.g. Burrage and Pohl [4]) The test problems used in our numerical comparisons differ with respect to computing costs of f evaluations and stability requirements. 2 Explicit pseudo two step RK methods We consider explicit pseudo two step RK methods (EPTRK methods) which have been recently introduced and investigated by ....
K. Burrage and B. Pohl, Implementing an ODE code on distributed memory computers, Computers Math. Applic. 28 (1994), 235-252.
On the Performance of Parallel Waveform Relaxations for.. - Burrage, Dyke (1995) Self-citation (Burrage Pohl) (Correct)
.... Department of Mathematics The University of Queensland Queensland 4072, Australia and Bert Pohl Seminar fur Angewandte Mathematik ETH Zurich, Ch 8092 Zurich, Switzerland Abstract In this paper, modifications to the distributed memory waveform relaxation package developed by Burrage and Pohl [2], known as PWVODE, are considered. Various dynamic, adaptive and communication protocols are considered, and a number of numerical comparisons on a 96 node Intel Paragon are presented which show the efficacy of PWVODE. 1 Introduction Consider solving systems of initial values problems (IVPs) of ....
....the possibility that the general waveform relaxation techniques may be slow to converge, methods of increasing the rate of convergence are sought. The multi splitting technique of O Leary and White [14] is discussed in section 3. Implementation details of the package developed by Burrage and Pohl [2] are considered in section 4. Modifications to the package are introduced in the context of a two dimensional test example in section 5. Finally, some conclusions are presented. 2 Waveform Relaxation Methods The system presented in (1) is generally nonlinear, and extremely stiff. When ....
[Article contains additional citation context not shown here]
K. Burrage and B. Pohl, Implementing an ODE code on distributed memory computers, Computers Math. Applic., 28, (1994), 235--252.
Parallel Methods for Systems of Ordinary Differential Equations - Burrage (1995) Self-citation (Burrage) (Correct)
....a 16 Kbyte cache. Only one processor per node is accessible for computation, and this processor has a peak rating of 75 Mflops. Before presenting the results we briefly discuss a number of issues relating to the effectiveness of any parallel wavefrom implementation. 3. 3 Ordering Burrage and Pohl [6] observed that there are two natural ways of ordering the components of the two dimensional Brusselator. The first ordering is u 11 ; Delta Delta Delta ; u 1N ; u 21 ; Delta Delta Delta ; u 2N ; Delta Delta Delta ; uNN ; v 11 ; Delta Delta Delta ; v 1N ; v 21 ; Delta Delta Delta ; ....
K. Burrage and B. Pohl, Implementing an ODE code on distributed memory computers, Computers Math. Applic., Vol. 28, No. 10--12, 1994, pp. 235--252.
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