Michael Sevilla: Virtualizing Resources for Big Data

Tuesday, June 5, 2012
E2-375

Michael will be talking about hist work on virtualizing system resources for big data processing. It will mainly reiterate the motivations and goals of the project. There will also be a discussion of the initial steps that we've made towards designing and implementing this new architecture.

The talk will touch on the current landscape of the available trace tools and the reasons for choosing different trace tools for this project. LTTng-2.0 and OProfile examples will be presented and analyzed to determine which events are important and to collect information about these events.