Loading Events

Map and Zoom Performance Methodology for Microservices

August 19 @ 3:30 pm - 5:00 pm UTC-7


Speakers: Harshad Sane and Kshitij Doshi (Intel)

Decomposition of monolithic applications into a collection of loosely coupled microservices simplifies solution architecture and accelerates deployment. By focusing optimization at a finer scope and by targeting each microservice independently, solution developers hope to simplify management of performance objectives. In practice however, microservice architecture gives rise to new issues in identifying performance and scaling hurdles – for example, chatty interactions between some microservices can increase overheads of communication and can hurt scaling when previously intra-process or intra-host messages now transit between processes/hosts, and through additional software abstractions of cloud services. Due to the variety and complex nature of compute resources that cloud providers offer, a disciplined methodology for performance analysis becomes critical.

In this tutorial, we describe a methodical approach for addressing the above issue and describe how to pursue multiple objectives such as debugging performance regressions, identifying scaling bottlenecks, exploring opportunities for optimization, etc. A critical first step is to verify health of different components and configurations. This is followed by a combination of iterating between two types of actions: that of mapping out performance/scaling properties at one plane of interactions, and that of progressively zooming into where the mapping identifies performance or scaling barriers.

Harshad Sane is a performance engineer in the Data Center and AI group with a deep technical expertise in system software, memory, and CPU architectures. He specializes in performance monitoring, software optimization, and tool development with focus in the cloud domain. Harshad joined Intel in 2008 after completing his undergraduate degree in Electrical and Telecommunications from College of Engineering and Technology (COET), Pune, India and M.S. in Electrical and Computer Engineering from UC Boulder.

Kshitij Doshi works at Intel Corporation in the Data Center and AI group, where he focuses on performance optimization of workloads and cloud instances. He obtained his undergraduate degree in Electrical Engineering from IIT Mumbai (1982) and his M.S. and Ph.D. degrees in Electrical and Computer Engineering from Rice University (1985 and 1989, respectively). His research interests span distributed systems, memory and storage architectures, and resource management.