Peter J. Haas
Communications in Statistics. Part C: Stochastic Models
Large-scale machine learning (ML) algorithms are often iterative, using repeated read-only data access and I/Obound matrix-vector multiplications to converge to an optimal model. It is crucial for performance to fit the data into single-node or distributed main memory and enable very fast matrix-vector operations on in-memory data. Generalpurpose, heavy- and lightweight compression techniques struggle to achieve both good compression ratios and fast decompression speed to enable block-wise uncompressed operations. Compressed linear algebra (CLA) avoids these problems by applying lightweight lossless database compression techniques to matrices and then executing linear algebra operations such as matrix-vector multiplication directly on the compressed representations. The key ingredients are effective column compression schemes, cache-conscious operations, and an efficient sampling-based compression algorithm. Experiments on an initial implementation in SystemML show in-memory operations performance close to the uncompressed case and good compression ratios.We thereby obtain significant end-to-end performance improvements up to 26x or reduced memory requirements.
Peter J. Haas
Communications in Statistics. Part C: Stochastic Models
Brian Hentschel, Peter J. Haas, et al.
EDBT 2018
Rainer Gemulla, Peter J. Haas, et al.
KDD 2011
Peter J. Haas, Gerald S. Shedler
Communications in Statistics. Stochastic Models