Large-scale computational grids consume substantial amount of energy due to their massive
sizes. The energy requirement for these grids for providing the power and cooling is becoming
comparable to the cost of acquisition. While a vast number of tools are available for service
providing managers to schedule jobs on grids, there is a lack of generally applicable methods for
reducing energy consumption while ensuring good quality of service.
Optimization of performance and energy consumption for large-scale grids requires developing
proper scheduling mechanisms that can support a variety of workflows. Grid workflows can
execute over hours or days and consists of large tasks, with or without precedence constraints.
The user or application is generally interested in the energy and performance characteristics of
the entire workflow. Since energy and performance are inversely proportional via a non-linear
relationship, workflow scheduling to meet the timing requirements while economizing energy is a
dual objective optimization problem. This is further accentuated by the fact that different classes
of users and/or their applications may have different priorities and require diverse energyperformance
tradeoffs. An effective resource management system that supports this also has to
be able to exploit the natural heterogeneous and adaptive nature of the grid resources.
GridPac (Grid with Power-Aware Computing) is a middleware environment
that will schedule multiple workflows across a distributed grid for system-wide optimization.
GridPac will be based on a novel framework for simultaneous optimizing of energy and