Abstract
Modern cloud-based applications adhere to a microservices architecture, encompassing numerous components interconnected through intricate dependencies, operating within a distributed environment. This talk gives an overview of three recent research projects aimed at solving practical challenges in promptly identifying and diagnosing outages.
The first paper employs an extreme event regularizer to predict outages within microservices. The second paper expounds on the construction of a Casual Graph for containerized microservices. The third paper introduces an approach leveraging time-series metric data and unstructured text from historical outage reports for outage root cause analysis. Furthermore, an overview of research themes at Adobe Research, Bangalore, will be presented.
Biography
Shiv is a Principal Research Scientist at Adobe Research, Bangalore. His current research centers around the domain of System Reliability and Efficiency. He uses and develops Causal Inference and Time Series tools to build early warning and diagnosis systems for cloud services. In the past, he has also delved into Marketing Attribution and Automated Insights domains. His work at Adobe Research has resulted in over 15 publications in top conferences, over 40 issued and filed patents, and over 40 technologies in Adobe products.
Beyond his personal research pursuits. Shiv leads Enterprise System Research agenda for Adobe Research. The main focus of his group revolves around the application of Machine Learning to enhance System Reliability and Efficiency, with particular emphasis on Cloud systems and Generative Models. Shiv holds a PhD from the University of Wisconsin at Madison.