Triangle counting is a building block for a wide range of graph applications. Traditional wisdom suggests that i) hashing is not suitable for triangle counting, ii) edge-centric triangle counting beats vertex-centric design, and iii) communication-free and workload balanced graph partitioning is a grand challenge for triangle counting. On the contrary, we advocate that i) hashing can help the key operations for scalable triangle counting on Graphics Processing Units (GPUs), i.e., list intersection and graph partitioning, ii) vertex-centric design reduces both hash table construction cost and memory consumption, which is limited on GPUs. In addition, iii) we exploit graph and workload collaborative, and hashing-based 2D partitioning to scale vertex-centric triangle counting over 1000 GPUs with sustained scalability. In this article, we present Trust which performs tr iangle co u nting with the ha s h operation and ver t ex-centric mechanism at the core. To the best of our knowledge, Trust is the first work that achieves over one trillion Traversed Edges Per Second (TEPS) rate for triangle counting.

中文翻译：

---

相信：在GPU上重新加载了三角计数

三角计数是各种图形应用程序的基础。传统观点认为，i）哈希不适用于三角计数； ii）以边缘为中心的三角计数胜过以顶点为中心的设计，并且iii）无通信和工作负载平衡的图分区是三角计数的一大挑战。相反，我们提倡：i）哈希可以帮助图形处理单元（GPU）上的可伸缩三角形计数的关键操作，即列表交集和图形分区，ii）以顶点为中心的设计可以降低哈希表的构建成本和内存消耗，这在GPU上受到限制。此外，iii）我们利用图和工作负载协作以及基于散列的2D分区来扩展具有1000个GPU的顶点为中心的三角形，并具有持续的可扩展性。在本文中，我们介绍相信 哪个执行 TR 伊朗克公司 ü 与哈 s h操作和版本 Ť 以核心为中心。据我们所知，相信 是第一个完成的工作 一万亿 三角形计数的每秒穿越边（TEPS）速率。