A5073365909- Distributed and Parallel Computing Systems
- Parallel Computing and Optimization Techniques
- Advanced Data Storage Technologies
- Scientific Computing and Data Management
- Industrial Vision Systems and Defect Detection
- Color Science and Applications
- Embedded Systems Design Techniques
- Distributed systems and fault tolerance
- Cloud Computing and Resource Management
- Stochastic Gradient Optimization Techniques
Supply Chain Competence Center (Germany)
2011-2016
Heidelberg University
2006-2007
Load balancing, maintenance, and energy efficiency are key challenges for upcoming supercomputers. An indispensable tool the accomplishment of these tasks is ability to migrate applications during runtime. Especially in HPC, where any performance hit frowned upon, such migration mechanisms have come with minimal overhead. This constraint usually not met by current practice adding further abstraction layers software stack. In this paper, we propose a concept MPI processes communicating over...
The eeClust project aims at reducing the energy consumption of applications on a HPC cluster by an integrated approach analysis, efficient management hardware power-states and monitoring clusters power consumption. application is traced trace file analyzed - manually with Vampir automatically Scalasca to determine phases in non-optimal utilization. source-code then instrumented API calls control daemon which switches runtime. This aware shared resources (e.g. network interface) only resource...