TERAFLUX · Exploiting dataflow parallelism in Teradevice Computing

Dataflow parallelism is key to reach power efficiency, reliability, efficient parallel programmability, scalability, data bandwidth. In this project we propose dataflow both at task level and inside thethreads, to offload and manage accelerated codes, to localize the computation, for managing the fault information with appropriate protocols, to easily migrate code to the available/working componentsand to respect the power/performance/temperature/reliability envelope, to efficiently handle the parallelism and have an easy and powerful execution model, to produce a more predictable behavior.While parallel systems have been around for many years, they were usually programmed and tuned by experts. In the future large scale systems will be widely available and therefore exploiting efficientlythe available parallelism will have to be easy enough to be accessible by the common user. Traditional programming models are either not very efficient for every application (message passing) or difficult toscale (shared memory). In order to address the programmability challenge we propose the use of a compiler directive based model to support an underlying dataflow-based thread execution that is known to exploit well the available parallelism and to efficiently move around large amounts of data. In particular we propose to use a model that offersdataflow scheduling of parallel execution threads. Combining multithreading with dataflow allows to exploit the available parallelism without the overheads of the original dataflow techniques.The multithreading dataflow model is expected to perform well for a number of classes of applications.An important contribution is provided by prof. Gao's team, who has been developing dataflow concepts for decades and has joined the TERAFLUX project after its initial phase.TERAFLUX is now bringing together top experts in dataflow in both continents Europe and Americas, with the aim to reach the higher goal of demonstrating for the first time the efficiency dataflow concept for the Exascale parallel computers of the 2020 and beyond.