Is your performance analysis approach as cutting edge as your application architecture?

To analyze and improve the performance of modern applications, you must abandon outdated approaches and toolsets which are rooted to the physical topology of servers and JVMs. Jon Hodgson discusses a new paradigm to reveal unexpected relationships and hotspots obscured by the elasticity of containers and microservices so that you can find and fix issues with the most overarching business impact.

With the adoption of elastic computing, containers, and microservices, application topology is often transient, nebulous, and unpredictably interrelated with countless other applications and services. This has posed a serious challenge for performance monitoring, since many tools and techniques are still rooted to static physical concepts like tiers, servers, and JVMs. The nature of performance problems has similarly increased in complexity. It is no longer sufficient to simply identify the “most overloaded servers” or “top slowest code.” A new performance analysis paradigm is needed that can identify the overarching issues that have the greatest impact on business goals and revenue. Jon Hodgson presents a new performance analysis paradigm and covers real-world case studies where these techniques revealed major problems which previously went undiagnosed. In one case, the techniques uncovered a seemingly small issue that had an unexpectedly large cascading effect across dozens of applications and affected millions of transactions for tens of thousands of users each day. One centralized fix dramatically improved end-user experience and the bottom line for multiple lines of business. This session is sponsored by Riverbed.