A survey on checkpointing strategies: Should we always checkpoint à la Young/Daly?

L Bautista-Gomez, A Benoit, S Di, T Herault… - Future Generation …, 2024 - Elsevier
Abstract The Young/Daly formula provides an approximation of the optimal checkpointing
period for a parallel application executing on a supercomputing platform. It was originally …

SLATE: Design of a modern distributed and accelerated linear algebra library

M Gates, J Kurzak, A Charara, A YarKhan… - Proceedings of the …, 2019 - dl.acm.org
The SLATE (Software for Linear Algebra Targeting Exascale) library is being developed to
provide fundamental dense linear algebra capabilities for current and upcoming distributed …

Accelerating geostatistical modeling and prediction with mixed-precision computations: A high-productivity approach with parsec

S Abdulah, Q Cao, Y Pei, G Bosilca… - … on Parallel and …, 2021 - ieeexplore.ieee.org
Geostatistical modeling, one of the prime motivating applications for exascale computing, is
a technique for predicting desired quantities from geographically distributed data, based on …

Extreme-scale task-based cholesky factorization toward climate and weather prediction applications

Q Cao, Y Pei, K Akbudak, A Mikhalev… - Proceedings of the …, 2020 - dl.acm.org
Climate and weather can be predicted statistically via geospatial Maximum Likelihood
Estimates (MLE), as an alternative to running large ensembles of forward models. The MLE …

Celerity: High-level c++ for accelerator clusters

P Thoman, P Salzmann, B Cosenza… - Euro-Par 2019: Parallel …, 2019 - Springer
In the face of ever-slowing single-thread performance growth for CPUs, the scientific and
engineering communities increasingly turn to accelerator parallelization to tackle growing …

Evolution of the SLATE linear algebra library

M Gates, A Abdelfattah, K Akbudak… - … Journal of High …, 2025 - journals.sagepub.com
SLATE (Software for Linear Algebra Targeting Exascale) is a distributed, dense linear
algebra library targeting both CPU-only and GPU-accelerated systems, developed over the …

Towards extreme scale technologies and accelerators for eurohpc hw/sw supercomputing applications for exascale: the textarossa approach

G Agosta, M Aldinucci, C Alvarez, R Ammendola… - Microprocessors and …, 2022 - Elsevier
In the near future, Exascale systems will need to bridge three technology gaps to achieve
high performance while remaining under tight power constraints: energy efficiency and …

Scaling implicit parallelism via dynamic control replication

M Bauer, W Lee, E Slaughter, Z Jia… - Proceedings of the 26th …, 2021 - dl.acm.org
We present dynamic control replication, a run-time program analysis that enables scalable
execution of implicitly parallel programs on large machines through a distributed and …

A 2D hydrodynamic model for shallow water flows with significant infiltration losses

Y Ni, Z Cao, Q Liu, Q Liu - Hydrological Processes, 2020 - Wiley Online Library
Infiltration losses may be significant and warrant proper incorporation into mathematical
models for river floods in arid and semi‐arid areas, rainfall‐induced surface runoffs in …

Exploiting data sparsity for large-scale matrix computations

K Akbudak, H Ltaief, A Mikhalev, A Charara… - … Conference on Parallel …, 2018 - Springer
Exploiting data sparsity in dense matrices is an algorithmic bridge between architectures
that are increasingly memory-austere on a per-core basis and extreme-scale applications. In …