Fault Tolerant High Performance Computing

Información del proyecto

FTHPC

Identificador del acuerdo de subvención: 818252

DOI

10.3030/818252

Proyecto cerrado

Fecha de la firma de la CE 17 Diciembre 2018

Fecha de inicio 1 Junio 2019

Fecha de finalización 31 Mayo 2025

Financiado con arreglo a

• EXCELLENT SCIENCE - European Research Council (ERC)

Coste total

€ 1 824 467,00

Aportación de la UE

€ 1 824 467,00

Coordinado por

THE HEBREW UNIVERSITY OF JERUSALEM
Israel

Periodic Reporting for period 4 - FTHPC (Fault Tolerant High Performance Computing)

Período documentado: 2023-12-01 hasta 2025-05-31

Resumen del contexto y de los objetivos generales del proyecto

The future of super-computing depends on our ability to cope with errors. Increasing machine size and decreasing operating voltage generate soft errors (bit flips) and hard errors (component failure). Hardware trends imply at least two errors per minute on next generation (exascale) supercomputers. If a CPU or GPU breaks down, the computed results must remain trustworthy. The high performance computing and the theory of computing communities have been addressing this challenge for more than two decades. Most solutions are either
(i) general purpose, requiring little to no algorithmic effort, but severely degrade performance (e.g. checkpoint-restart), or (ii) tailored to specific applications, very efficient, but significantly increasing programmers’ workload.
We propose a novel approach that achieves the best of both worlds: high performance and general purpose fault resilience. Objectives include: (1) Pushing the limits of resilience-enabling building blocks. (2) Developing low-overhead algorithmic tools that are as widely applicable as possible. (3) Improving usability and exposing automating opportunities.

Trabajo realizado desde el comienzo del proyecto hasta el final del período abarcado por el informe y los principales resultados hasta la fecha

We obtained fault tolerant parallel matrix multiplication algorithms that reduce the resource overhead by minimizing both the number of additional processors and the communication costs. We obtained fault resilience with small overhead costs, for Strassen's and other recursive fast matrix multiplication algorithms. We propose a new straggler mitigation solution for delay faults, and apply it to distributed matrix multiplication.
We demonstrated that computing over encoded data can serve a dual purpose: mitigating faults on the one hand, and improving resource utilization on the other hand, including computation costs, communication costs, while reducing memory footprint. We generalized this approach and applied it to several algorithms and implementations. Including for example long integer multiplication. Some of the techniques allow for better performance compared to the original algorithms.

Avances que van más allá del estado de la técnica e impacto potencial esperado (incluida la repercusión socioeconómica y las implicaciones sociales más amplias del proyecto hasta la fecha)

Our algorithmic solutions allow for faults tolerance, while providing high performance, that no previous state-of-the-art solution can guarantee. Furthermore, the encoded computation yields the fastest matrix multiplication implementations. E.g our solutions significantly outperform Intel's dgemm, the state of the art for CPU and Nvidia's CUDA on GPU.

idt-picture-for-ec-portal.png