Research
Research Interests
I am a researcher in the field of high performance computing with a focus on data compression and sparse matrix storage. My research interests include: HPC, data compression, sparse matrix storage, sparse algorithms, CUDA programming, and performance portability. I am currently working on the development of high performance compression algorithms for the FZ Project and the IVSparse Library. I have also had experience as an undergraduate research intern at Sandia National Laboratories working on the E3SM atmospheric model, HOMME (High-Order Method Modeling Environment).
Projects
FZ
The FZ Project is a lossy error-bound lossy compression framework funded by the NSF. These compressors see use in many fields such as AI, quantum simulation, scientific computing, and instrumentation.
I am currently working on the development team of the IU FZ Team under the advisement of Dr. Fengguang Song. Thus far I’ve developed the FZModules library, a modular framework for constructing bespoke lossy compression pipelines. FZModules is designed for use by domain scientists and compression researchers to develop compression pipelines for specific applications.
IVSparse (Primary Co-Contributor)
The IVSparse Library is a C++ sparse matrix library optimized for the compression of sparse data in cases where non-zero values are highly redundant. IVSparse supports three main compression formats.
- CSC: Compressed Sparse Column - The standard sparse matrix format that stores the column indices of non-zero values and the values themselves.
- VCSC: Value Compressed Sparse Column - A format that takes advantage of the redundancy in the values of the matrix by storing unique values and their counts.
- ~2.25x fold compression over CSC (for redundant data)
- IVCSC: Indexed Value Compressed Sparse Column - A format that further compresses the index storage of VCSC by using positive-delta encoding and then bytepacking indices for each unique value in a column.
- ~7.5x fold compression over CSC (for redundant data)
Publications
-
[SC Workshops ‘25 (DRBSD)] Skyler Ruiter, Jiannan Tian, Fengguang Song. “FZModules: A Heterogenous Computing Framework for Customizable Data Compression Pipelines.” 2025 IEEE/ACM The 11th International Workshop on Data Analysis and Reduction for Big Scientific Data (DRBSD). St Louis, MO, USA. November 16-21. arxiv
-
[BigData ‘24] Seth Wolfgang, Skyler Ruiter, Marc Tunnell, Timothy Triche Jr, Erin Carrier, Zachary DeBruine. “Value-Compressed Sparse Column (VCSC): Sparse Matrix Storage for Single-cell Omics Data.” 2024 IEEE International Conference on Big Data (BigData). Washington, DC, USA. December 15-18. arxiv
Talks and Pictures
