Parallel Clustering Coefficient Computation using GPUs
نویسندگان
چکیده
Clustering coefficient is the measure of how tightly vertices are bounded in a network. The Triangle Counting problem is at the core of clustering coefficient computation. We present a new technique for implementing clustering coefficient algorithm on GPUs. It relies on neighbour list being sorted with respect to vertex ID. The algorithm can process very large graphs not seen in the literature for single-node in-memory systems before. Our technique is able to compute clustering coefficient of each vertex in power-law graphs with up to 512M edges on a single GPU. The GPU implementation offers 7x speedup over the best known work for the same graph. For the CPU implementation, we present results for graphs with up to 4B edges. We also investigate performance bottleneck of power-law graphs with skewed vertex degree distribution by analyzing GPU performance counter outputs and interpreting them in terms of GPU memory and thread characteristics.
منابع مشابه
Hierarchical Clustering with CUDA/GPU
Graphics processing units (GPUs) are powerful computational devices tailored towards the needs of the 3-D gaming industry for high-performance, real-time graphics engines. Nvidia Corporation provides a programming language called CUDA for general-purpose GPU programming. Hierarchical clustering is a common method used to determine clusters of similar data points in multidimensional spaces; if t...
متن کاملGPU-accelerated Gaussian clustering for fMPE discriminative training
The Graphics Processing Unit (GPU) has extended its applications from its original graphic rendering to more general scientific computation. Through massive parallelization, state-ofthe-art GPUs can deliver 200 billion floating-point operations per second (0.2 TFLOPS) on a single consumer-priced graphics card. This paper describes our attempt in leveraging GPUs for efficient HMM model training....
متن کاملDesign Space Exploration for GPU-Based Architecture
Recent advances in Graphics Processing Units (GPUs) provide opportunities to exploit GPUs for non-graphics applications. Scientific computation is inherently parallel, which is a good candidate to utilize the computing power of GPUs. This report investigates QR factorization, which is an important building block of scientific computation. We analyze different mapping mtheods of QR factorization...
متن کاملParallel Computations for Hierarchical Agglomerative Clustering using CUDA Fast and Scalable Computations on Graphics Processors
Graphics Processing Units (GPU) in today’s desktops can well be thought of as a high performance parallel processor. Traditionally, parallel computing is the usage of multiple computing resources to execute computational problems simultaneously. Such computations are possible using multi-core CPUs or computers with multiple CPUs or by using a network of computers in parallel. Today’s GPUs are c...
متن کاملAccelerating high-order WENO schemes using two heterogeneous GPUs
A double-GPU code is developed to accelerate WENO schemes. The test problem is a compressible viscous flow. The convective terms are discretized using third- to ninth-order WENO schemes and the viscous terms are discretized by the standard fourth-order central scheme. The code written in CUDA programming language is developed by modifying a single-GPU code. The OpenMP library is used for parall...
متن کامل