|
|
|
George Candea
|
|
Dependable Systems Laboratory
|
Assistant Professor PhD (Stanford University, 2005)
|
office(s):
INN330
phone(s): [+41 21 69] 34648,31432
fax: 021 693-8130
|
|
|
BIOGRAPHY
|
George Candea is an Assistant Professor of Computer Science at EPFL, heading the Dependable Systems Lab. His research is focused on techniques and tools for engineering reliable, highly available, manageable computer systems. Until 2009, he was also Chief Scientist of Aster Data, a large-scale data analytics company he co-founded in 2005. George received the Top 35 Young Technology Innovators award from the MIT Technology Review in 2005. In 2001, George was part of the founding team of the Recovery-Oriented Computing (ROC) project. Previously, he worked at Oracle Corp. in the server technologies division and held intern positions at IBM Research and Microsoft Research.
George received his PhD in computer science from Stanford University in 2005 and his BS (1997) and MEng (1998) in computer science from the Massachusetts Institute of Technology.
For more details, see Curriculum Vitae (PDF).
|
|
CURRENT WORK
|
|
George's research focuses primarily on tools, techniques, and runtimes that improve the dependability of software systems while also increasing programmer productivity. He seeks pragmatic solutions targeted primarily at large-scale software with millions of lines of code running in highly concurrent systems with hundreds of threads. Current projects fall into three main categories:
|
|
SERVICE
|
Recent program committees: OSDI 2010, ASPLOS 2010, DSN 2010, SOSP 2009 WIP, EuroSys 2009, USENIX 2009, HotOS 2009, DSN 2009, TRANSACT 2008, DSN 2008, ICDCS 2007, HotOS 2007, USENIX 2007
|
|
MAIN PUBLICATIONS
|
Deadlock Immunity: Enabling Systems To Defend Against Deadlocks, Horatiu Jula, Daniel Tralamazza, Cristian Zamfir, George Candea, 8th Symp. on Operating Systems Design and Implementation (OSDI), San Diego, CA, December 2008
Microreboot - A Technique for Cheap Recovery, George Candea, Shinichi Kawamoto, Yuichi Fujiki, Greg Friedman, Armando Fox, 6th Symposium on Operating Systems Design and Implementation (OSDI), San Francisco, CA, December 2004
A Scalable, Predictable Join Operator for Highly Concurrent Data Warehouses, George Candea, Neoklis Polyzotis, Radek Vingralek, 35th Intl. Conf. on Very Large Data Bases (VLDB), Lyon, France, August 2009
Middleware-based Database Replication: The Gaps Between Theory and Practice, Emmanuel Cecchet, George Candea, Anastasia Ailamaki, ACM SIGMOD International Conference on Management of Data (SIGMOD), Vancouver (BC), Canada, 2008
LFI: A Practical and General Library-Level Fault Injector, Paul D. Marinescu and George Candea, Intl. Conf. on Dependable Systems and Networks (DSN), Lisbon, Portugal, June 2009
ConfErr: A Tool for Assessing Resilience to Human Configuration Errors, Lorenzo Keller, Prasang Upadhyaya, George Candea, Intl. Conf. on Dependable Systems and Networks (DSN), Anchorage, AK, June 2008
Reverse-Engineering Drivers for Safety and Portability, Vitaly Chipounov and George Candea, 4th Workshop on Hot Topics in System Dependability (HotDep), San Diego, CA, December 2008
Crash-Only Software, George Candea and Armando Fox, 9th Workshop on Hot Topics in Operating Systems (HotOS), Lihue, Hawaii, May 2003
Reducing Recovery Time in a Small Recursively Restartable System, George Candea, James Cutler, Armando Fox, Rushabh Doshi, Priyank Garg, Rakesh Gowda, Intl. Conf. on Dependable Systems and Networks (DSN), Washington, D.C., June 2002
Recursive Restartability: Turning the Reboot Sledgehammer into a Scalpel, George Candea and Armando Fox, 8th Workshop on Hot Topics in Operating Systems (HotOS), Schloss Elmau, Germany, May 2001 Complete list of publications available at http://dslab.epfl.ch/pubs
|
|
TEACHING
|
|
|
|