Registration - It's Free, Join Now

Scaling Up a Real Application on Azul

In this technical session, developers will be taken on an ‘E’ ticket ride on the topic of scaling today’s transaction-intensive applications. This session offers a short case study of tuning and scaling a Java™ technology-based application. Cliff starts with tuning a “black box” application, which is an ordinary device driver model checking application - this is a good candidate for tuning since it uses many small bits of work or independent tasks. Doing any one task may produce more tasks. The work is highly unstructured and very fine grained. In theory, this coding style should be endlessly scalable but in practice there were a number of issues, mostly due to the fine-grained nature of the tasks. The initial attempt used only a handful of CPUs inside an Azul Compute Appliance, but after three or four small modifications, scalability was cranked up to all 384 CPUs - and ran the job 45x faster than the Intel®-based server upon which it was running. Time to complete a proof dropped from *weeks* to *hours.*