CONPAR 1986: CONPAR 86 pp 104-111

# Multitasking algorithms on CRAY computers for interval arithmetic Newton-like methods for a class of systems of nonlinear equations

• Hartmut Schwandt
Namertal Algorithms (Session 1.2)
Part of the Lecture Notes in Computer Science book series (LNCS, volume 237)

## Abstract

For several classes of systems of nonlinear equations interval arithmetic methods can be defined which converge to a solution essentially under the condition that an initial inclusion is known, i.e. the convergence can be said to be global. We consider vector algorithms for an interval arithmetic Newton-like method which is combined with an interval arithmetic “fast solver” for nonlinear systems with block tridiagonal Jacobian. As an example we consider a nine point discretization of a twodimensional nonlinear Dirichlet problem. More specifically we discuss efficient algorithms for multiprocessor computers with two and four vector processors. The algorithms are intended to use all processors in parallel as far as possible under the condition that the vector efficiency in the parallel tasks does not substantially decrease. We report numerical results for programs using the multitasking routines on 1 and 2-processor CRAY-X/MP.

## Keywords

Interval Arithmetic Parallel Task Interval Vector Synchronization Point Interval Operation

## References

1. [1]
B.L. Buzbee, G.H. Golub, C.W. Nielson: On DirectMethods for Solving Poisson's Equation, SIAM J.Num.Anal. 7(1970), 627–656.Google Scholar
2. [2]
S.S. Chen, J.J. Dongarra, C.C. Hsiung: Multiprocessing for linear algebra algorithms on the CRAY X-MP-2: Experience with small granularity, J.Par.Distr.Comp. 1(1984), 22–31.Google Scholar
3. [3]
CRAY Multitasking User's Guide, Ref. SN-0222 B, CRAY Research, Mendota Heights, 1986.Google Scholar
4. [4]
R.W. Hockney: (r,n1/2,s1/2) measurement on the 2-CPU CRAY X-MP, Par.Comp. 2(1985), 1–14.Google Scholar
5. [5]
J. Ortega, W.C. Rheinboldt: Iterative Solution of Nonlinear Equations in Several Variables, Academic Press, New York, 1970.Google Scholar
6. [6]
H.Schwandt: Almost globally convergent interval methods for discretizations of nonlinear elliptic partial differential equations, to appear in SIAM J.Num.Anal.Google Scholar
7. [7]
Schwandt, H.: Newton-like interval methods for large systems of nonlinear equations on vector computers, Comp. Phys. Comm. 37(1985), 223–232.Google Scholar
8. [8]
H. Schwandt: An interval arithmetic approach for the construction of an almost globally convergent method for the solution of the nonlinear Poisson equation on the unit square, SIAM J.Sc.St.Comp. 5(1984), 427–452.Google Scholar
9. [9]
H. Schwandt: The solution of nonlinear elliptic Dirichlet problems on rectangles by almost globally convergent interval methods, SIAM J.Sc.St.Comp., 6(1985), 617–638.Google Scholar