Numerical algorithms and fault tolerance of hyperexascale computer systems

DOI10.1134/S1064562417010021MaRDI QIDQ2358586zbMATH OpenFDO

Authors M. V. Jakobovskij, B. N. Chetverushkin

Publication date 15 June 2017

Published in Doklady Mathematics (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1134/S1064562417010021

Mathematics Subject Classification ID

PDEs in connection with computer science (35Q68) Computer aspects of numerical algorithms (65Y99) Reliability, testing and fault tolerance of networks and computer systems (68M15) Performance evaluation, queueing, and scheduling in the context of computer systems (68M20)

Recommendations

• Fault tolerant algorithms for heat transfer problems
• Is the multigrid method fault tolerant? The multilevel case
• Numerical recovery strategies for parallel resilient Krylov linear solvers.
• Is the multigrid method fault tolerant? the two-grid case
• Resilience for massively parallel multigrid solvers

Cites work

• Estimates of the difference between approximate solutions of the Cauchy problems for the parabolic diffusion equation and a hyperbolic equation with a small parameter
• Explicit schemes and numerical simulation using ultrahigh-performance computer systems
• Hyperbolic type explicit kinetic scheme of magneto gas dynamics for high performance computing systems
• Kinetic schemes and a quasi-gas-dynamic system of equations
• On conservation laws for hyperbolized equations
• Simulating flows of incompressible and weakly compressible fluids on multicore hybrid computer systems

Cited in

(3)

• Numerical algorithms for systems with extramassive parallelism
• Parallel computing, failure recovery, and extreme values
• Conservation laws and a compact quasi-gasdynamic system

This page was built for publication: Numerical algorithms and fault tolerance of hyperexascale computer systems

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2358586)