8 papers
Stencil-Lifting: Hierarchical Recursive Lifting System for Extracting Summary of Stencil Kernel in Legacy Codes
Mingyi Li, Junmin Xiao, Siyan Chen +5
We introduce Stencil-Lifting, a novel system for automatically converting stencil kernels written in low-level languages in legacy code into semantically equivalent Domain-Specific…
SparStencil: Retargeting Sparse Tensor Cores to Scientific Stencil Computations via Structured Sparsity Transformation
Qi Li, Kun Li, Haozhi Han +7
Sparse Tensor Cores offer exceptional performance gains for AI workloads by exploiting structured 2:4 sparsity. However, their potential remains untapped for core scientific worklo…
An Accurate and Efficient Large-scale Regression Method through Best Friend Clustering
Kun Li, Liang Yuan, Yunquan Zhang +1
As the data size in Machine Learning fields grows exponentially, it is inevitable to accelerate the computation by utilizing the ever-growing large number of available cores provid…
An Efficient Vectorization Scheme for Stencil Computation
Kun Li, Liang Yuan, Yunquan Zhang +3
Stencil computation is one of the most important kernels in various scientific and engineering applications. A variety of work has focused on vectorization and tiling techniques, a…
Reducing Redundancy in Data Organization and Arithmetic Calculation for Stencil Computations
Kun Li, Liang Yuan, Yunquan Zhang +3
Stencil computation is one of the most important kernels in various scientific and engineering applications. A variety of work has focused on vectorization techniques, aiming at ex…
AutoFlow: Hotspot-Aware, Dynamic Load Balancing for Distributed Stream Processing
Pengqi Lu, Liang Yuan, Yunquan Zhang +2
Stream applications are widely deployed on the cloud. While modern distributed streaming systems like Flink and Spark Streaming can schedule and execute them efficiently, streaming…