Batched BLAS Operations at SIAM CSE17

Over the last year, there has been significant interest in solving many small linear algebra problems simultaneously. Library vendors such as MKL and NVIDIA, along with researchers at instutions including Manchester, Tennessee, and Sandia National Labs have all been attempting to perform these calculations as efficiently as possible.

Over the weekend prior to the SIAM CSE17 meeting, many of those researchers (including myself) held a workshop to discuss strategies for batched BLAS (basic linear algebra subprogram) computations. Furthermore, lots of discussion was aimed at standardising the function APIs and the memory layout that users will interact with. The slides, and a number of research papers on the topic, are available at this page.

At the SIAM CSE17 meeting, our team at Manchester organised a minisymposium to discuss the highlights of our weekend with a wider audience. A brief summary of the four talks, along with a copy of their slides, is given below.

Using implicit matrices in Python

There are lots of new features in SciPy 0.13 (release notes) but for me the most important are the updated matrix functions in scipy.linalg and the one norm estimator in scipy.sparse.linalg.

In some of my recent research (related to section 4 of this) I’ve needed to estimate the one norm of a large (n^4 x n^2) dense matrix without computing each element. All we can assume is the ability to compute matrix-vector products (via some rather complicated function), meaning we only know the entries of the matrix implicitly.