"When things go wrong in today’s systems," writes Brown CS PhD student Jonathan Mace, "it can be difficult to answer questions about causes of failures, uncover dependencies between components, or understand performance or resource usage." Very few things are going wrong for Jonathan at the moment, nor is his recent performance difficult to understand: he's just followed a Best Paper Award at SOSP with earning a Facebook Graduate Fellowship. It's part of a program that provides funding to support the research of promising Computer Science and Engineering PhD students.
"We are impressed with your research," writes Facebook's Yee Jiun Song, "and how you are impacting the field." Currently, Jonathan's research focuses on how to effectively monitor and troubleshoot distributed systems at runtime. In particular, his recent work on Pivot Tracing (co-created with alum Ryan Roelke and Brown CS Assistant Professor Rodrigo Fonseca) presents abstractions for cross-component monitoring. It combines dynamic instrumentation with causal tracing techniques, and enables users to capture and correlate arbitrary statistics across application, component, and machine boundaries without needing to change or redeploy system code.
The fellowship covers Jonathan's tuition and fees for nine months of the academic year(s) 2016-2017 and 2017-2018, provides a $37,000 stipend per year, and pays for a visit to Facebook headquarters to present his research and meet with colleagues in the field.