Adaptive step-size control is a critical feature for the robust and efficient numerical
solution of initial-value problems in ordinary differential equations. In this
paper, we show that adaptive step-size control can be incorporated within a
family of parallel time integrators known as revisionist integral deferred
correction (RIDC) methods. The RIDC framework allows for various strategies to
implement step-size control, and we report results from exploring a few of
them.
Keywords
initial-value problems, revisionist integral deferred
correction, parallel time integrators, local error
estimation, adaptive step-size control
Mathematical Subject Classification 2010
Primary: 65H10, 65L05, 65Y05
Milestones
Received: 9 October 2013
Revised: 10 December 2014
Accepted: 17 December 2014
Published: 27 March 2015
Authors
Andrew J. Christlieb
Department of Mathematics
Michigan State University
East Lansing, 48823
United States