We have developed an architecture for Continuous Program Optimization (CPO) to assist in, and automate the challenging task of performance tuning a system. CPO utilizes the data from all hardware and software layers to detect, diagnose, and eliminate performance problems. We describe the CPO architecture and several CPO agent implementations that we used do demonstrate the capabilities of this framework.
Bio: Dr. Calin Cascaval is a research scientist and manager of the Programming Models and Tools for Scalable Systems at IBM Research. He received his Ph.D. degree in Computer Science from the University of Illinois at Urbana-Champaign in 2000. He joined IBM Research the same year and has worked on system software, simulators and compilers for the BlueGene projects. He was the team lead for compiler research in the and PERCS (DARPA HPCS) project. His interests are in programming models and languages for scalable systems, performance analysis, optimization tools and compilers for large parallel systems, and the interface between compilers and architectures. For more details and a list of his publications , please visit his homepage at http://www.research.ibm.com/people/c/cascaval.