Analytical modeling for what-if analysis in complex cloud computing applications.

Modern cloud applications are complex distributed systems with tens or hundreds of interacting software components. An important management task in cloud computing platforms is to predict the impact of a certain workload or reconfiguration change on the performance of the application. Such predictions require the design of "what-if" models of the application that take as input hypothetical changes in the application's workload or environment and estimate its impact on performance. We present a workload-based what-if analysis system that uses commonly available monitoring information in large scale systems to enable the administrators to ask a variety of workload-based "what-if" queries about the system. We use a network of queues to analytically model the behavior of large distributed cloud applications. Our system automatically generates node-level queueing models and then uses model composition to build system-wide models. We employ a simple what-if query language and an intelligent query execution algorithm that employs on-the-fly model construction and a change propagation algorithm to efficiently answer queries on large scale systems. We have built a prototype and have used traces from two large production cloud bapplications from a financial institution as well as real-world synthetic applications to evaluate its what-if modeling framework. Our experimental evaluation validates the accuracy of our node-level resource usage, latency and workload models and then shows how our system enables what-if analysis in four different cloud applications.