A Self Distributing Virtual Machine for Adaptive Multicore Environments

摘要

The use of parallel systems is no longer limited to dedicated clusters as multicore chips are more and more appearing in embedded applications. To meet power, performance and cost targets these systems need to be adaptive. The reconfiguration features of recent FPGAs make new approaches for this type of parallel computing possible: Dynamic reconfiguration at runtime offers an important step to adaptive behavior of systems-on-chip (SoCs). This article analyzes the challenges of such an adaptive SoC. It is shown that many of the requirements for an adaptive FPGA-realization are met by the SDVM, the scalable dataflow-driven virtual machine which has been successfully implemented and tested on a cluster of workstations. The SDVM has evolved to a virtualization layer for multicore-FPGAs, now called SDVM~R. This virtualization layer allows a transparent runtime-reconfiguration of the underlying hardware to adapt to the changing system environment. Results for a basic application for both systems are presented.