Improving Scheduling Techniques in Heterogeneous Systems With Dynamic, On-Line Optimisations

Provided by: University of Bielefeld
Topic: Hardware
Format: PDF
Computational performance increasingly depends on parallelism, and many systems rely on heterogeneous resources such as GPUs and FPGAs to accelerate computationally intensive applications. However, implementations for such heterogeneous systems are often hand-crafted and optimized to one computation scenario, and it can be challenging to maintain high performance when application parameters change. In this paper, the authors demonstrate that machine learning can help to dynamically choose parameters for task scheduling and load-balancing based on changing characteristics of the incoming workload.

Find By Topic