Abstract The performance of three implementations of the restarted GMRES algorithm with randomized-SVD-based preconditioning has been analyzed. They have been tested with a wide population of matrices of varying properties and compared to the standard ILU(0) preconditioning. The implementations comprise two variants of the preconditioner, one of them implemented for CPU-only and hybrid GPU–CPU executions to better assess the benefits and pitfalls in both contexts. The trade-off between iteration-to-solution and time-to-solution metrics is discussed and it is shown that a competitive convergence rate is attained. In addition, the CPU off-loading to the GPU leads to a significant improvement of the second metric for the largest matrices analyzed. Such an in efficiency is promising considering the various codes and tools that employ the restarted GMRES algorithm.

Abstract Every computation presents a small chance that an unexpected phenomenon ruins or modifies its output. Computers are prone to errors that, although may be very unlikely, are hard, expensive or simply impossible to avoid. In the exascale, with thousands of processors involved in a single computation, those errors are especially harmful because they can corrupt or distort the results, wasting human and material resources. In the present work, we study the effect of ionizing radiation on several pieces of commercial hardware, very common in modern supercomputers. Aiming to reproduce the natural radiation that could arise, CPUs (Xeon, EPYC) and GPUs (A100, V100, T4) are subject to a known flux of neutrons coming from two radioactive sources, namely $$^{252}$$ 252 Cf and $$^{241}$$ 241 Am-Be, in a special irradiation facility. The working hardware is irradiated under supervision to quantify any appearing error. Once the hardware response is characterised, we are able to scale down the radiation intensity and to estimate the effects on standard data centres. This can help administrators and researchers to develop their contingency plans and protocols.

ISDEP (Integrator of Stochastic Differential Equations for Plasmas) is a Monte Carlo code that solves the plasma dynamics in a fusion device and perfectly scales on distributed computing platforms. Montera is a recent framework developed for achieving Grid efficient executions of Monte Carlo applications, as ISDEP is. In this work, the improvement of performing the calculations of ISDEP with Montera, which rise up to 34.9%, is shown as well as an analysis on the implications it could have, which aim to show to the fusion research community the benefits of using Montera.