Towards network-aware Data Mining

Distributed data mining algorithms executing on a shared network of workstations often suffer from unpre-dictable performance problems due to limited network re-sources that are being shared. We show that data min-ing algorithms, which have an approximate nature, can adapt to network-resource constraints. We argue that exist-ing network monitoring and quality of service mechanisms are insufficient to support the needs of such applications. We then discuss the mechanisms (systems support) needed to support such network-aware applications. Finally, we describe the current status (work-in-progress) of the sys-tem under development and discuss some general issues involved in developing resource-aware data mining algo-rithms