Iterative Algorithms for Stochastically Robust Static Resource Allocation in Periodic Sensor Driven Clusters

V. Shestak, J. Smith, A.A. Maciejewski, and H.J. Siegel (USA)


Heterogeneous systems, resource allocation, stochastic op timization, iterative algorithms, computer clusters.


This research investigates the problem of robust static re source allocation for a large class of clusters processing periodically updated data sets under an imposed quality of service constraint. The target hardware platform consists of a number of sensors generating input for heterogeneous ap plications continuously executing on a set of heterogeneous compute nodes. In practice such systems are expected to function in a physical environment replete with uncertainty, which causes the amount of processing required over time to fluctuate substantially. Determining a resource alloca tion that accounts for this uncertainty in a way that can pro vide a probabilistic guarantee that a given level of QoS is achieved is an important research problem. The stochastic robustness metric is based on a mathematical model where the relationship between uncertainty in system parameters and its impact on system performance is described stochas tically. The established metric is then used in the design of several resource allocation algorithms utilizing evolu tionary approaches. The performance results and compari son analysis are presented for a simulated environment that replicates a heterogeneous cluster-based processing center for a radar system.

Important Links:

Go Back