mLR: Scalable Laminography Reconstruction based on Memoization

TL;DR

This paper introduces mLR, a scalable laminography reconstruction method that employs memoization to significantly reduce computation time and memory usage, enabling large-scale 3D reconstructions on limited hardware.

Contribution

mLR innovatively applies memoization to ADMM-FFT, making laminography reconstruction more scalable and efficient for large input sizes across multiple GPUs.

Findings

Achieved 52.8% average performance improvement over ADMM-FFT.

Enabled laminography reconstruction on 2Kx2Kx2K input problems.

Reduced memory consumption and computation time significantly.

Abstract

ADMM-FFT is an iterative method with high reconstruction accuracy for laminography but suffers from excessive computation time and large memory consumption. We introduce mLR, which employs memoization to replace the time-consuming Fast Fourier Transform (FFT) operations based on an unique observation that similar FFT operations appear in iterations of ADMM-FFT. We introduce a series of techniques to make the application of memoization to ADMM-FFT performance-beneficial and scalable. We also introduce variable offloading to save CPU memory and scale ADMM-FFT across GPUs within and across nodes. Using mLR, we are able to scale ADMM-FFT on an input problem of 2Kx2Kx2K, which is the largest input problem laminography reconstruction has ever worked on with the ADMM-FFT solution on limited memory; mLR brings 52.8% performance improvement on average (up to 65.4%), compared to the original ADMM-FFT.