Community detection is the problem of identifying natural divisions in networks. Efficient parallel algorithms for identifying such divisions is critical in a number of applications, where the size of datasets have reached significant scales. This technical report presents an optimized parallel implementation of Leiden, a high quality community detection method, for shared memory multicore systems. On a server equipped with dual 16-core Intel Xeon Gold 6226R processors, our Leiden implementation, which we term as GVE-Leiden, outperforms the original Leiden, igraph Leiden, and NetworKit Leiden by 373x, 86x, and 7.2x respectively - achieving a processing rate of 352M edges/s on a 3.8B edge graph. Compared to GVE-Louvain, our parallel Louvain implementation, GVE-Leiden achieves an 11x reduction in disconnected communities, with only a 36% increase in runtime. In addition, GVE-Leiden improves performance at an average rate of 1.6x for every doubling of threads.

翻译：社区检测是识别网络中自然划分的问题。在数据集规模已达到显著量级的诸多应用中，高效并行识别此类划分的算法至关重要。本技术报告提出了一种针对共享内存多核系统优化的高质量社区检测方法——Leiden的并行实现。在配备双路16核Intel Xeon Gold 6226R处理器的服务器上，我们的Leiden实现（称为GVE-Leiden）较原始Leiden、igraph Leiden和NetworKit Leiden分别实现了373倍、86倍和7.2倍的性能提升——在含38亿条边的图上达到每秒3.52亿条边的处理速率。与我们的并行Louvain实现GVE-Louvain相比，GVE-Leiden将不连通社区数量减少了11倍，而运行时间仅增加36%。此外，每增加一倍线程数，GVE-Leiden性能平均提升1.6倍。