The expressiveness of different parallel languages and paradigms provide natural ways to solve different clases of problems. It would be an advantage to be able to link different parallel modules or libraries written using different parallel languages.

Our work on this area includes the development of Converse, an interoperable framework which extracts the essential aspects of parallel runtime support into a set of core components. Languages developed on top of Converse include PVM and MPI.

Recent work includes implementing library APIs commonly used in scientific applications on top of a Charm++ based framework. This would allow applications written using those APIs to run with minimal modification while gaining the benefits of virtualization and load balancing. Projects in this category includes Adaptive MPI. A partial implementation of ARMCI is also in the works.