04/07/2025
By Danielle Fretwell

The Francis College of Engineering, Department of Electrical and Computer Engineering, invites you to attend a Doctoral Dissertation defense by Jiaxi Chen on: "Optimization Of DCT-Based Lossy Compression Algorithm For Scientific Datasets."

Candidate Name: Jiaxi Chen
Degree: Doctoral
Defense Date: Wednesday, April 9th, 2025
Time: 1– 2:30 p.m.
Location: Ball Hall, Room 302

Committee:
Advisor: Seung Woo Son, Ph.D., Associate Professor, Electrical & Computer Engineering, UMass Lowell

Committee Members
1. Xuejun Lu, Ph.D., Professor, Electrical & Computer Engineering, UMass Lowell
2. Yan Luo, Ph.D., Professor, Computer Engineering, UMass Lowell

Brief Abstract:
High-performance computing (HPC) systems that run scientific simulations of significance produce a large amount of data during runtime. Transferring or storing such big datasets causes a severe I/O bottleneck and a considerable storage burden. One approach to mitigate such overhead is to reduce the size of data by compression techniques before transferring it through the network or storing it on disk. Traditional lossless compression can preserve all the information with full fidelity but is not able to achieve appreciable data reduction. Unlike lossless compression algorithms, error-controlled lossy compressors could significantly reduce the data size while respecting the user-defined error bound. Though lossy compression has been studied for years, it is not widely used in the scientific domain to date. Several reasons are summarized as follows. Firstly, scientific datasets require guaranteeing the user-defined error bound rather than the visual data compression by human eyes, in other words, a higher compression accuracy is needed. Secondly, most lossy compressors work with single-precision floating point numbers, thus double-precision numbers compression needs to be studied.
This dissertation focuses on the optimization of DCTZ, one of the transform-based lossy compressors with a highly efficient encoding and purpose-built error control mechanism that accomplishes high compression ratios with high data fidelity. DCTZ reduces the size of the input data by preserving only small parts of coefficients in the frequency domain. However, since DCTZ quantizes the DCT coefficients in the frequency domain, it may only partially control the relative error bound defined by the user. This dissertation investigates the performance of DCTZ with real-world scientific datasets, improves the compression quality, and proposes a new mode of DCTZ.