On Efficient Constructions of Checkpoints

Yu Chen · Zhenming LIU · Bin Ren · Xin Jin

Keywords: [ Systems and Software ] [ General Machine Learning Techniques ]

[ Abstract ] [ Join Zoom
[ Slides
Please do not share or post zoom links


Efficient construction of checkpoints/snapshots is a critical tool for training and diagnosing deep learning models. In this paper, we propose a lossy compression scheme for checkpoint constructions (called LC-Checkpoint). LC-Checkpoint simultaneously maximizes the compression rate and optimizes the recovery speed, under the assumption that SGD is used to train the model. LC-Checkpoint uses quantization and priority promotion to store the most crucial information for SGD to recover, and then uses a Huffman coding to leverage the non-uniform distribution of the gradient scales. Our extensive experiments show that LC-Checkpoint achieves a compression rate up to 28× and recovery speedup up to 5.77× over a state-of-the-art algorithm (SCAR).

Chat is not available.