Computing rank-revealing factorizations of matrices stored out-of-core
This work addresses the bottleneck of handling large-scale matrices in data-intensive applications like scientific computing or machine learning, offering incremental improvements over traditional methods for out-of-core storage.
The paper tackles the problem of computing rank-revealing factorizations for matrices too large to fit in RAM by developing efficient algorithms that minimize communication costs when data is stored on external memory devices. It introduces two methods, including a blocked column pivoted QR and a UTV factorization with power iterations, which achieve computational times scaling as cn^3 with only marginally larger constants compared to in-memory processing.
This paper describes efficient algorithms for computing rank-revealing factorizations of matrices that are too large to fit in RAM, and must instead be stored on slow external memory devices such as solid-state or spinning disk hard drives (out-of-core or out-of-memory). Traditional algorithms for computing rank revealing factorizations, such as the column pivoted QR factorization, or techniques for computing a full singular value decomposition of a matrix, are very communication intensive. They are naturally expressed as a sequence of matrix-vector operations, which become prohibitively expensive when data is not available in main memory. Randomization allows these methods to be reformulated so that large contiguous blocks of the matrix can be processed in bulk. The paper describes two distinct methods. The first is a blocked version of column pivoted Householder QR, organized as a "left-looking" method to minimize the number of write operations (which are more expensive than read operations on a spinning disk drive). The second method results in a so called UTV factorization which expresses a matrix $A$ as $A = U T V^*$ where $U$ and $V$ are unitary, and $T$ is triangular. This method is organized as an algorithm-by-blocks, in which floating point operations overlap read and write operations. The second method incorporates power iterations, and is exceptionally good at revealing the numerical rank; it can often be used as a substitute for a full singular value decomposition. Numerical experiments demonstrate that the new algorithms are almost as fast when processing data stored on a hard drive as traditional algorithms are for data stored in main memory. To be precise, the computational time for fully factorizing an $n\times n$ matrix scales as $cn^{3}$, with a scaling constant $c$ that is only marginally larger when the matrix is stored out of core.