This paper presents the influence of the fault tolerance configuration on different applications using performance metrics. Two configuration parameters are analysed: the heartbeat/watchdog interval and the checkpoint interval. In addition, even message logging is mandatory, an analysis of its overhead on different applications is presented. The impact of message logging on applications has been analysed according to the nature of the communication primitives used on the application. This analysis shows why for different applications the message logging introduces different overhead.Presentado en el IX Workshop Procesamiento Distribuido y Paralelo (WPDP)Red de Universidades con Carreras en Informática (RedUNCI
The demand for computational power has been leading the improvement of the High Performance Computin...
bouteill,lemarini,gk,fci lri.fr MPI is one of the most adopted programming models for Large Cluste...
Abstract. Using rollback-recovery based fault tolerance (FT) techniques in applications executed on ...
This paper presents the influence of the fault tolerance configuration on different applications usi...
Descripció del recurs: el 23 de febrer de 2010¿Es adecuado un sistema rápido pero poco robusto?¿Es a...
The increasing failure rate in High Performance Computing encourages the investigation of fault tole...
In High Performance Computing (HPC) the demand for more performance is satisfied by increasing the n...
AbstractThe increasing failure rate in High Performance Computing encourages the investigation of fa...
Researchers have mentioned that the three most difficult and growing problems in the future of high-...
Researchers have mentioned that the three most difficult and growing problems in the future of high-...
Fault tolerance can allow processes executing in a computer system to survive failures within the sy...
Abstract—We present an analysis design of how to incorpo-rate a transparent fault tolerance system a...
International audienceFault tolerance in MPI becomes a main issue in the HPC community. Several appr...
Fault tolerance has become an important issue for parallel applications in the last few years. The p...
La tendencia general de los computadores paralelos es crecer en complejidad y en número de component...
The demand for computational power has been leading the improvement of the High Performance Computin...
bouteill,lemarini,gk,fci lri.fr MPI is one of the most adopted programming models for Large Cluste...
Abstract. Using rollback-recovery based fault tolerance (FT) techniques in applications executed on ...
This paper presents the influence of the fault tolerance configuration on different applications usi...
Descripció del recurs: el 23 de febrer de 2010¿Es adecuado un sistema rápido pero poco robusto?¿Es a...
The increasing failure rate in High Performance Computing encourages the investigation of fault tole...
In High Performance Computing (HPC) the demand for more performance is satisfied by increasing the n...
AbstractThe increasing failure rate in High Performance Computing encourages the investigation of fa...
Researchers have mentioned that the three most difficult and growing problems in the future of high-...
Researchers have mentioned that the three most difficult and growing problems in the future of high-...
Fault tolerance can allow processes executing in a computer system to survive failures within the sy...
Abstract—We present an analysis design of how to incorpo-rate a transparent fault tolerance system a...
International audienceFault tolerance in MPI becomes a main issue in the HPC community. Several appr...
Fault tolerance has become an important issue for parallel applications in the last few years. The p...
La tendencia general de los computadores paralelos es crecer en complejidad y en número de component...
The demand for computational power has been leading the improvement of the High Performance Computin...
bouteill,lemarini,gk,fci lri.fr MPI is one of the most adopted programming models for Large Cluste...
Abstract. Using rollback-recovery based fault tolerance (FT) techniques in applications executed on ...