Bandwidth-aware Co-allocating Meta-schedulers for Mini-grid Architectures DRAFT

In this paper, we present several bandwidth-aware co-allocating meta-schedulers that take into account inter-cluster network utilization as a means by which to mitigate the slowdown associated with the interaction of simultane-ously co-allocated jobs in a dedicated computational grid. We make use of a bandwidth-centric parallel job commu-nication model that captures the time-varying utilization of shared inter-cluster network resources. By doing so, we are able to evaluate the performance of grid scheduling algo-rithms that focus not only on node resource allocation, but also on shared inter-cluster network bandwidth.