Scaling the "Memory Wall" for Multi-Dimensional Seismic Processing with Algebraic Compression on Cerebras CS-2 Systems

We exploit the high memory bandwidth of AI-customized Cerebras CS-2 systems for seismic processing. By leveraging low-rank matrix approximation, we fit memory-hungry seismic applications onto memory-austere SRAM wafer-scale hardware, thus addressing a challenge arising in many wave-equation-based al...

Full description

Saved in:
Bibliographic Details
Published in:International Conference for High Performance Computing, Networking, Storage and Analysis (Online) pp. 1 - 12
Main Authors: Ltaief, Hatem, Hong, Yuxi, Wilson, Leighton, Jacquelin, Mathias, Ravasi, Matteo, Keyes, David
Format: Conference Proceeding
Language:English
Published: ACM 11.11.2023
Subjects:
AI-optimized Architecture
Bandwidth
Convolution
Deconvolution
Energy Efficiency
Extreme Parallelism
Frequency-domain analysis
High Memory Bandwidth
High performance computing
Low-Carbon Energy Applications
Low-Rank Matrix Approximation
Memory management
Random access memory
Seismic Processing
ISSN:2167-4337
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:We exploit the high memory bandwidth of AI-customized Cerebras CS-2 systems for seismic processing. By leveraging low-rank matrix approximation, we fit memory-hungry seismic applications onto memory-austere SRAM wafer-scale hardware, thus addressing a challenge arising in many wave-equation-based algorithms that rely on Multi-Dimensional Convolution (MDC) operators. Exploiting sparsity inherent in seismic data in the frequency domain, we implement embarrassingly parallel tile low-rank matrix-vector multiplications (TLR-MVM), which account for most of the elapsed time in MDC operations, to successfully solve the Multi-Dimensional Deconvolution (MDD) inverse problem. By reducing memory footprint along with arithmetic complexity, we fit a standard seismic benchmark dataset into the small local memories of Cerebras processing elements. Deploying TLR-MVM execution onto 48 CS-2 systems in support of MDD gives a sustained memory bandwidth of 92.58PB/s on 35,784,000 processing elements, a significant mile-stone that highlights the capabilities of AI-customized architectures to enable a new generation of seismic algorithms that will empower multiple technologies of our low-carbon future.
ISSN:2167-4337
DOI:10.1145/3581784.3627042