PERFORMANCE ANALYSIS IN HIGH-SPEED WIDE AREA IP-OVER-ATM NETWORKS - TOP-TO-BOTTOM END-TO-END MONITORING

We describe an approach to the analysis of the performance of distribu ted applications in high-speed wide area networks. The approach is des igned to identify all of the issues that impact performance, and isola te the causes due to the related hardware and software components. We also describe the use of a distributed parallel data server as a netwo rk load generator that can be used in conjunction with this approach t o probe various aspects of high-speed distributed systems from top to bottom of the protocol stock and from end to end in the network. To de monstrate the utility of this approach we present the analysis of a TC P-over-ATM problem that was uncovered while developing this methodolog y. This work was done in conjunction with the ARPA-funded MAGIC gigabi t testbed.