Dr. Emre Kıcıman

Talk

Fluxo: A Simple Service Compiler

Over the last 10-15 years, our industry has developed and deployed many large-scale Internet services, from e-commerce to social networking sites, all facing common challenges in performance, reliability, and scalability. To address these challenges, developers consistently draw from a relatively small repertoire of software architecture design patterns or best practices, such as replication, tiering, pre-computation, and caching. The application of these techniques across different services, however, varies significantly in both their design and implementation because their effectiveness depends heavily on the service?s semantic requirements, workloads, performance and other runtime and environmental characteristics.
Most of this critical information is directly measurable, and, in this presentation, discuss our research towards building a system, analogous to a profile-driven optimizing compiler, that separates architectural decisions to support performance, reliability, and scalability from service functionality. Through three mechanisms, our tool, Fluxo, separates an Internet service?s logical functionality from the architectural decisions made to support performance, scalability, and reliability: 1) a coarse-grained dataflow-based programming model; 2) detailed runtime request tracing to capture workload distributions, performance behavior, and resource requirements; and 3) a set of analysis techniques that determine how to apply simple, parameterized dataflow transformations to optimize the service architecture for performance, scalability, and reliability.

Biography

Emre Kıcıman is a researcher in the Internet Services Research Center (ISRC) at Microsoft Research, where his interests are broadly in the area of large-scale Internet services, their operations, and their end-to-end reliability. His work focuses on ways to improve monitoring and analysis of system behaviors to improve their design, reliability and performance. Most recently, he's been working on Web application performance and debugging, as well as dabbling in applications of search and social networking.