The DLB (Dynamic Load Balancing) library and LeWl (LEnd When Idle) algorithm provide a runtime solution to deal with the load imbalance of parallel applications independently of the source of imbalance. DLB relies on the usage of hybrid programming models and exploits the malleability of the second level of parallelism to redistribute computation power across processes.; When executing real applications with LeWl, although application's performance is significantly improved, we have observed in some cases efficiency values between 60% and 70%, far from our theoretical limit.; This work is a deep analysis of the sources of efficiency loss correlated with application characteristics, parallelization schemes and programming models. We have bas...
Scientific applications often contain large and computationally- intensive parallel loops. Dynamic l...
Load balance is critical for performance in large parallel applica-tions. An imbalance on today’s fa...
Scientific applications are often complex, irregular, and computationally-intensive. To accommodate ...
It is well known that load imbalance is a major source of efficiency loss in HPC (High Performance C...
In parallel iterative applications, computational efficiency is essential for addressing large probl...
Parallel iterative applications often suffer from load imbalance, one of the most critical performan...
In parallel computing, obtaining maximal performance is often mandatory to solve large and complex p...
Load imbalance is a long-standing source of inefficiency in high performance computing. The situati...
A non-preemptive hybrid load balancing algorithm is proposed for heterogeneous distributed computing...
This research demonstrates that the automatic implementation of a dynamic load balancing (DLB) strat...
The hybrid programming model MPI+OpenMP are useful to solve the problems of load balancing of parall...
International audienceThe performance of irregular scientific applications can be easily affected by...
Abstract. Traditional load balancing algorithms for data-intensive iterative routines can successful...
A large class of computational problems are characterised by frequent synchronisation, and computati...
High performance parallel and distributed computing systems are used to solve large, complex, and da...
Scientific applications often contain large and computationally- intensive parallel loops. Dynamic l...
Load balance is critical for performance in large parallel applica-tions. An imbalance on today’s fa...
Scientific applications are often complex, irregular, and computationally-intensive. To accommodate ...
It is well known that load imbalance is a major source of efficiency loss in HPC (High Performance C...
In parallel iterative applications, computational efficiency is essential for addressing large probl...
Parallel iterative applications often suffer from load imbalance, one of the most critical performan...
In parallel computing, obtaining maximal performance is often mandatory to solve large and complex p...
Load imbalance is a long-standing source of inefficiency in high performance computing. The situati...
A non-preemptive hybrid load balancing algorithm is proposed for heterogeneous distributed computing...
This research demonstrates that the automatic implementation of a dynamic load balancing (DLB) strat...
The hybrid programming model MPI+OpenMP are useful to solve the problems of load balancing of parall...
International audienceThe performance of irregular scientific applications can be easily affected by...
Abstract. Traditional load balancing algorithms for data-intensive iterative routines can successful...
A large class of computational problems are characterised by frequent synchronisation, and computati...
High performance parallel and distributed computing systems are used to solve large, complex, and da...
Scientific applications often contain large and computationally- intensive parallel loops. Dynamic l...
Load balance is critical for performance in large parallel applica-tions. An imbalance on today’s fa...
Scientific applications are often complex, irregular, and computationally-intensive. To accommodate ...