Tiling is a key technique for data locality optimization and is widely used in high-
performance implementations of dense matrix-matrix multiplication for multicore/manycore …

Today's Internet Services are undergoing fundamental changes and shifting to an intelligent
computing era where AI is widely employed to augment services. In this context, many …

This work optimizes tensor-times-dense matrix multiply (Ttm) for general sparse and semi-
sparse tensors on CPU and NVIDIA GPU platforms. Ttm is a computational kernel in tensor …

Deep neural networks have been employed in a broad range of applications, including face
detection, natural language processing, and autonomous driving. Yet, the neural networks …

SpMV, the product of a sparse matrix and a dense vector, is emblematic of a new class of
applications that are memory bandwidth and communication, not flop, driven. Sparsity and …

SpMV is an essential kernel existing in many HPC and data center applications. Meanwhile,
the emerging many-core hardware provides promising computational power, and is widely …

Modern optimizing compilers are able to exploit memory access or computation patterns to
generate vectorization codes. However, such patterns in irregular applications are unknown …

Abstract Graphic Processing Units (GPUs) have become popular in the last decade due to
their high memory bandwidth and powerful computing capacity. Nevertheless, achieving …