Connie Yee

Senior Data Scientist at Bloomberg

Connie Yee currently builds data science models on financial datasets at Bloomberg LP and has previously worked on NLP tools for news and social media data at Thomson Reuters. She has over 10 years of industry and research experience in developing software and machine learning algorithms. Connie holds a Master’s degree in Operations Research from Columbia University and Bachelor’s degree from MIT.

WATCH LIVE: February 16 @ 4:00PM – 4:30PM ET

Monolithic Code Analysis Using Graph Theory

Over time, monolithic software systems have tendency to become overgrown and tightly integrated. One way to improve the overall architecture is to migrate the monolith to newer paradigms, in particular using microservices. The challenge lies in determining which functionality to decouple from the intertwined and connected code, and how to weave them into the grand migration plan. We will present an approach using graph theory to identify areas of a monolithic application that can break dependencies and decompose it into modular monoliths. We will demonstrate how graph theory concepts, such as centrality and minimum node cut, can provide insights to help drive decisions for a migration strategy.