Proactive Performance Engineering in a DevOps Context Prior Year Content

DevOps methodologies require stronger cohesion between application development and operations teams. In the DevOps paradigm, performance engineering must mature beyond the usual measure-and-tune approach. Join Sundar Narayanan as he explores principles and techniques for proactive performance engineering. Sundar shares techniques for building predictive performance models from system designs parameterized with workload characteristics and expected resource demands. Using operational measurement data from profiling and monitoring tools, these models are refined continuously to increase their accuracy. These models enable the development team to proactively predict performance and scalability challenges before they happen. Sundar also describes techniques for early performance testing and optimization with standardized instrumentation and measurement frameworks.