The FP7 research project SAVE has been successfully completed. Over the last three years, Maxeler Technologies has joined a consortium of seven partners from both industry and academia to address the challenges of processing large, dynamic workloads on Heterogeneous System Architectures (HSA) with CPUs and dedicated Dataflow Engines (DFE) as well as other accelerators. The project has led to a number of innovations including novel OS components to support self-adaptive management of the HSA and virtualisation for the various types of compute resources. Maxeler has developed a new virtualisation layer for its multiscale dataflow computers, including a management and monitoring infrastructure and virtualisation support for DFEs. These innovations are highly relevant for running large-scale applications in multi-user HPC environments for moving toward cloud deployments of such applications.
“Self-adaptive behavior and dynamic resource allocation is key to improving performance and efficiency in heterogeneous HPC systems with dedicated DFEs. Adding orchestration and virtualisation to Maxeler’s dataflow computers helps us to efficiently run large dynamic workloads in a multitenant environment, a technology that is highly promising for a range of users in finance, science and engineering.” said Georgi Gaydadjiev, Vice President, Dataflow Software Engineering.
Full story here.