Hypergraph partitioning finds practical applications in various fields, such as high-performance computing and circuit partitioning in VLSI physical design, where high-performance solutions often demand substantial parallelism beyond what existing CPU-based solutions can offer. While GPUs are promising in this regard, their potential in hypergraph partitioning remains unexplored. In this work, we first develop an end-to-end deterministic hypergraph partitioner on GPUs, ported from state-of-the-art multi-threaded CPU work, and identify three major performance challenges by characterizing its performance. We propose the first end-to-end solution, gHyPart , to unleash the potentials of hypergraph partitioning on GPUs. To overcome the challenges of GPU thread underutilization due to imbalanced workload, long critical path, and high work complexity due to excessive operations, we redesign GPU algorithms with diverse parallelization strategies thus expanding optimization space; to address the challenge of no one-size-fits-all implementation for various input hypergraphs, we propose a decision tree-based strategy to choose a suitable parallelization strategy for each kernel. Evaluation on 500 hypergraphs shows up to 125.7× (17.5× on average), 640.0× (24.2× on average), and 171.6× (1.4× on average) speedups over two CPU partitioners and our GPU baseline gHyPart-B , respectively.

Point clouds are crucial for 3D geometry representation, and vital in applications like autonomous driving and augmented reality. Despite advancements in deep learning-based analytics, their high computational cost limits deployment on edge devices with constrained resources. To this end, we analyze PointNet++, a leading point cloud analytics framework, identifying two major bottlenecks: 1) GPU is underutilized due to limited parallelism and excessive kernel launches in the sampling stage and voting stage, and 2) irregular memory accesses in the grouping stage. To address these, we propose parallel sampling and voting to enhance GPU utilization and fuse subroutines in grouping to improve memory efficiency. Experimental results demonstrate that our optimizations result in significant speedup (up to $5.0 \times, 3.2 \times$ on average) across various point cloud workloads on edge devices.

Attracted by the enormous potentials of Graphics Processing Units (GPUs), an array of efforts has surged to deploy Breadth-First Search (BFS) on GPUs, which, however, often exploits the static mechanisms to address the challenges that are dynamic in nature. Such a mismatch prevents us from achieving the optimal performance for offloading graph traversal on GPUs. To this end, we propose XBFS that leverages the runtime optimizations atop GPUs to cope with the nondeterministic characteristics of BFS with the following three techniques: First, XBFS adaptively exploits four either new or optimized frontier queue generation designs to accommodate various BFS levels that present dissimilar features. Second, inspired by the observation that the workload associated with each vertex is not proportional to its degree in bottom-up, we design three new strategies to better balance the workload. Third, XBFS introduces the first truly asynchronous bottom-up traversal which allows BFS to visit vertices for multiple levels at a single iteration with both theoretical soundness and practical benefits. Taken together, XBFS is, on average, 3.5×, 4.9×, 11.2× and 6.1× faster than the state-of-the-art Enterprise, Tigr, Gunrock on a Quadro P6000 GPU and Ligra on a 24-core Intel Xeon Platinum 8175M CPU. Note, the CPU used for Ligra is more expensive than the GPU for XBFS.