Energy-conscious Scheduling of Parallel Applications Utilising Dynamic Voltage Scaling

Ref: 12111
A parallel program to run on heterogeneous multiprocessor platforms with the objective to minimize both total execution time and energy consumption.

Key advantages
  • Low energy consumption and good quality schedules at the same time
  • Low algorithmic (time) complexity leading to ease of implementation

Background

The invention

The invention adopts dynamic voltage scaling (DVS) to minimize energy consumption. DVS can be used with a number of recent commodity processors that are enabled to operate at different voltage supply levels at the expense of sacrificing clock frequencies.

In the context of scheduling, this multiple voltage facility implies that there is a trade-off between the quality of schedules and energy consumption. Our algorithm effectively balances these two performance goals using a novel optimization method, which takes into account both goals.

Applications

Resource management environments in high-performance computing systems (e.g., server farms in data centres, supercomputers and computer clusters).

Principal inventors