A parallel language and its programming system for heterogeneous networks

Abstract

SUMMARYThe paper presents a new parallel language, mpC, designed specially for programming high-performance computations on heterogeneous networks of computers, as well as its supportive programming environment. The main idea underlying mpC is that an mpC application explicitly defines an abstract network and distributes data, computations and communications over the network. The mpC programming environment uses, at run time, this information as well as information on any real executing network in order to map the application to the real network in such a way that ensures the efficient execution of the application on this real network. Experience of using mpC for solving both regular and irregular real-life problems on networks of heterogeneous computers is also presented.

HeteroMPI+ScaLAPACK: Towards a ScaLAPACK (Dense Linear Solvers) on Heterogeneous Networks of Computers

“…Both strategies were implemented in the mpC language [10,11]. The first strategy is implemented using calls to ScaLAPACK; the second strategy is implemented with calls to LAPACK and BLAS [12].…”

Parallel testing of distributed software

“…The test manager is implemented in the mpC language [6][7][8]. Therefore, it uses the relevant mpC means to detect the number of physical processors and their actual relative speed at run time.…”

Parallel Computing on Heterogeneous Networks

“…An original approach to parallel computing on heterogeneous networks that has been proposed and implemented in the framework of the mpC language [8][9] and the HMPI library [10] and their programming systems. In brief, this approach can be summarised as follows:…”