TEACHING ACTIVITY

Full time collaborator professor at the Computer Architecture Department (DAC) of the Technical University of Catalonia (UPC) assigned to the Barcelona School of Informatics (FIB).

Course 2006/2007 - currently

• Projecte de Sistemes Operatius (ProSO)

Former teaching activity

Formerly, besides FIB, I was assigned to other engineering schools in the UPC: the Telecommunication Engineering School of Barcelona (ETSETB) and the Engineering School of UPC at Vilanova i la Geltrú (EPSEVG). My former courses are:

• Course 2006/2007 (FIB)
• Sistemes Operatius (SO)
• Course 2005/2006 - 2006/2007 (ETSETB)
• Laboratori d'Arquitectura i Sistemes Operatius (LabAriso)
• Course 2005/2006 (ETSETB)
• Arquitectura i Sistemes Operatius II (ARISOII)
• Course 1998/1999 - 2004/2005 (EPSEVG)
• Introducció als Sistemes Operatius (ISOP)
• Sistemes Operatius (SIOP)

I'm a member of the "Performance management for next-generation data centers" working area, which is collocated in the Autonomic Systems and eBusiness Platforms of the Bacelona Supercomputing Center (BSC) and in the High Performance Computing Group at the Computer Architecture Department (DAC) in the UPC.

The field of my current research is resource management for heterogeneous workloads on virtualized execution environments. In particular, the main focus of my work is the on-line profiling of the applications to use in the prediction of their service level and to guide the resource allocation decisions.

More information can be found in www.bsc.es/autonomic and in the web page of the "Performance management for next-generation data centers" working area

Former reseach activity

Formerly, my research was focussed on resource management for Java applications. In my PhD dissertation we showed how these applications can benefit from the knowledge that the Java virtual machine has about the application behavior. We've implemented a memory prefetch technique based on the cooperation between the Java virtual machine and the operating system that significantly improve the performace of the memory management. In our proposal, the Java virtual machine implements a high accurated prediction about the use of memory and guides the operating system decisions about the memory management.