Novel runtime systems support for adaptive compositional modeling in PSEs

Problem solving environments (PSEs) have progressed significantly in the past few years. The vision of truly seamless PSEs relies on runtime systems support that is cognizant of the operational issues underlying scientific computations and, at the same time, is flexible enough to accommodate diverse application scenarios. This paper presents a PSE runtime support solution through a novel combination of two computational technologies – Weaves, a source-language independent parallel runtime compositional framework that operates through reverse-analysis of compiled object files, and runtime recommender systems that aid in dynamic knowledge-based application composition. Domain-specific adaptivity is exploited through runtime recommendation of code modules and a sophisticated checkpointing framework for transparent deployment. A core set of “adaptivity schemas” are provided as templates for adaptive composition of large-scale scientific computations. Implementation issues, motivating application contexts, and preliminary results are described.