Imperial College London
Alternate

ProfessorPaulKelly

Faculty of Engineering, Department of Computing

Professor of Software Technology
 
 
 
//

Contact

 

+44 (0)20 7594 8332p.kelly Website

 
 
//

Location

 

Level 3 (upstairs), William Penney Building, room 304William Penney LaboratorySouth Kensington Campus

//

Summary

 

Publications

Citation

BibTex format

@article{Kukreja:2020:10.5194/gmd-2020-325,
author = {Kukreja, N and Hückelheim, J and Louboutin, M and Washbourne, J and Kelly, PHJ and Gorman, GJ},
doi = {10.5194/gmd-2020-325},
journal = {Geoscientific Model Development},
title = {Lossy Checkpoint Compression in Full Waveform Inversion},
url = {http://dx.doi.org/10.5194/gmd-2020-325},
year = {2020}
}
Download

RIS format (EndNote, RefMan)

TY  - JOUR
AB  - This paper proposes a new method that combines check- pointing methods with error-controlled lossy compression for large-scale high-performance Full-Waveform Inversion (FWI), an inverse problem commonly used in geophysical exploration. This combination can signif- icantly reduce data movement, allowing a reduction in run time as well as peak memory.In the Exascale computing era, frequent data transfer (e.g., memory bandwidth, PCIe bandwidth for GPUs, or network) is the performance bottleneck rather than the peak FLOPS of the processing unit.Like many other adjoint-based optimization problems, FWI is costly in terms of the number of floating-point operations, large memory foot- print during backpropagation, and data transfer overheads. Past work for adjoint methods has developed checkpointing methods that reduce the peak memory requirements during backpropagation at the cost of additional floating-point computations.Combining this traditional checkpointing with error-controlled lossy compression, we explore the three-way tradeoff between memory, precision, and time to solution. We investigate how approximation errors introduced by lossy compression of the forward solution impact the objective function gradient and final inverted solution. Empirical results from these numerical experiments indicate that high lossy-compression rates (compression factors ranging up to 100) have a relatively minor impact on convergence rates and the quality of the final solution.
AU  - Kukreja,N
AU  - Hückelheim,J
AU  - Louboutin,M
AU  - Washbourne,J
AU  - Kelly,PHJ
AU  - Gorman,GJ
DO  - 10.5194/gmd-2020-325
PY  - 2020///
SN  - 1991-959X
TI  - Lossy Checkpoint Compression in Full Waveform Inversion
T2  - Geoscientific Model Development
UR  - http://dx.doi.org/10.5194/gmd-2020-325
UR  - https://gmd.copernicus.org/preprints/gmd-2020-325/
UR  - http://hdl.handle.net/10044/1/92272
ER  -
Download