Workshop on runtime and operating systems for supercomputers, Venice, Italy
25 June 2012 - 25 June 2012
Italy
A workshop on runtime and operating systems for supercomputers will take place from 25 to 29 June 2012 in Venice, Italy.
The complexity of node architectures in supercomputers is increasing as development progresses. Factors such as rising levels of parallelism in multi- and many-core chips, the emerging heterogeneity of computational resources, and energy and memory constraints, are forcing a re-evaluation of approaches towards operating systems and runtime environments.
The event will be a forum for researchers to exchange ideas and discuss research questions that are relevant to upcoming generation of supercomputers. Topics will include:
- operating systems and runtime system scalability on many-node and multi/many-core systems;
- specialised operating systems for Supercomputing;
- distributed/hybrid/partitioned operating systems and runtime systems for supercomputing;
- fault tolerance;
- system noise analysis and prevention;
- interaction between middleware, runtime system, and the operating system;
- resource management and forwarding;
- parallel job start-up;
- memory management and emerging memory technologies;
- real-time considerations for supercomputing.For further information, please visit: http://www.mcs.anl.gov/events/workshops/ross/2012/(opens in new window)
The complexity of node architectures in supercomputers is increasing as development progresses. Factors such as rising levels of parallelism in multi- and many-core chips, the emerging heterogeneity of computational resources, and energy and memory constraints, are forcing a re-evaluation of approaches towards operating systems and runtime environments.
The event will be a forum for researchers to exchange ideas and discuss research questions that are relevant to upcoming generation of supercomputers. Topics will include:
- operating systems and runtime system scalability on many-node and multi/many-core systems;
- specialised operating systems for Supercomputing;
- distributed/hybrid/partitioned operating systems and runtime systems for supercomputing;
- fault tolerance;
- system noise analysis and prevention;
- interaction between middleware, runtime system, and the operating system;
- resource management and forwarding;
- parallel job start-up;
- memory management and emerging memory technologies;
- real-time considerations for supercomputing.For further information, please visit: http://www.mcs.anl.gov/events/workshops/ross/2012/(opens in new window)