Some Gpu Algorithms for Graph Connected Components and Spanning Tree

Abstract

Graphics Processing Units (GPU) are application specific accelerators which provide high performance to cost ratio and are widely available and used, hence places them as a ubiquitous accelerator. A computing paradigm based on the same is the general purpose computing on the GPU (GPGPU) model. The GPU due to its graphics lineage is better suited for the data-parallel, data-regular algorithms. The hardware architecture of the GPU is not suitable for the data parallel but data irregular algorithms such as graph connected components and list ranking. In this paper, we present results that show how to use GPUs efficiently for graph algorithms which are known to have irregular data access patterns. We consider two fundamental graph problems: finding the connected components and finding a spanning tree. These two problems find applications in several graph theoretical problems. In this paper we arrive at efficient GPU implementations for the above two problems. The algorithms focus on minimising irregularity at both algorithmic and implementation level. Our implementation achieves a speedup of 11-16 times over a corresponding best sequential implementation.

Hybrid algorithms for list ranking and graph connected components

“…Similarly, our hybrid algorithm for graph connected components is faster by 25% compared to the best known GPU implementation [26].…”

“…On such regular applications, GPUs can outperform a single-core CPU performance by a large factor on average. In recent times, researchers have studied how GPUs perform on irregular computations such as list ranking [33], [21], connected components [26], among others. It is to be noted that in these cases, the speed-up compared to a single core CPU performance is only of the order of 10 or less.…”

Work efficient parallel algorithms for large graph exploration on emerging heterogeneous architectures

“…Popular parallel algorithms in the PRAM model include the algorithm of Shiloach and Vishkin [41] and its variants by Greiner [19]. On GPUs, a variant of Shiloach and Vishkin [41] is used by Soman et al [42]. A heterogeneous execution of this algorithm on a CPU+GPU platform with an improvement of 35% on average is shown in [6].…”

“…However, because of the irregular nature of operations involved, this workload is often difficult to implement on most modern parallel architectures. Efficient implementations of the Shiloach and Vishkin algorithm are known to exist for a variety of parallel architectures including symmetric multiprocessors [5], Cray and CM2 [19], GPUs [42], and also on CPU+GPU systems [6].…”

“…This reduces the thread divergence to a great extent and also reduces data movement. More details of these optimizations are described in [42].…”