Seminar Title: Simulation Based Capacity Planning for Web Search Engines
Speaker: Mauricio Marin, University of Santiago, Santiago, Chile.
Date/Time: July 27th, 2016 at 2:00 pm
Location: Room 3213 VS (VSim Building)
Abstract: Large scale Web search engines are complex and highly optimized systems devised to operate on dedicated clusters of processors. Any, even a small, gain in performance is beneficial to economical operation given the large amount of hardware resources deployed in the respective data centers. Performance is fully dependent on users behavior which is featured by unpredictable and drastic variations in trending topics and query arrival rate intensity. In this context, this talk describes different techniques devised to quickly predict performance on large scale Web search engines operating on dedicated clusters of processors. It focuses on the capacity planning problem by describing a method to predict the amount of data center resources required to serve a given query processing workload. The method is based on a combination between formulae from mean value operational analysis and what we call approximate parallel discrete-event simulation.
Short Bio of Speaker: Dr. Mauricio Marin is a former researcher at Yahoo! Labs Santiago hosted by the University of Chile, and currently a full professor at University of Santiago, Chile. He holds a PhD in Computer Science from University of Oxford, UK, and a MSc from University of Chile. His research work is on parallel computing and distributed systems with applications in query processing and capacity planning for Web search engines. He has obtained research grants on parallel query processing upon distributed metric-space databases and scalable parallel algorithms and data structures for text search and indexing.