Advanced Computing and Optimization Infrastructure for Extremely Large-Scale Graphs on Post Peta-Scale Supercomputers
Advanced Computing and Optimization Infrastructure for Extremely Large-Scale Graphs on Post Peta-Scale Supercomputers
复制标题
后 Peta 级超级计算机上超大规模图的高级计算和优化基础设施
DOI：
10.1007/978-981-13-1924-2_11
复制
发表时间：
2016
期刊：
影响因子：
--
通讯作者：
Toshio Endo
中科院分区：
文献类型：
--
作者：
Katsuki Fujisawa ;Toyotaro Suzumura;Hitoshi Sato;Koji Ueno;Yuichiro Yasui;Keita Iwabuchi;Toshio Endo
在本文中，我们介绍了我们正在进行的研究项目。高性能计算（HPC）领域的许多正在进行的研究项目的目标是在千万亿次超级计算机上开发用于超大规模图的高级计算和优化基础设施。最近在各种应用领域（如交通、社交网络、网络安全、防灾和生物信息学）中出现的超大规模图需要快速和可扩展的分析。Graph 500基准测试以每秒遍历边数（TEPS）衡量任何超级计算机执行广度优先搜索（BFS）的性能。在2014-2017年，我们的项目团队在K计算机上实现了约38.6TeraTEPS，并在第8和第10至第15届Graph 500基准测试中赢家。2013年开始了面向大规模城市的Urban OS（Operating System）开发研究项目。Urban OS被认为是信息物理系统（CPS）的新兴应用之一，通过利用传感器技术收集人员分布和交通运动的大数据集，并将其存储在云存储系统中。下一步，我们将应用优化、模拟和深度学习技术来解决这些问题，并检查在网络空间中获得的解决方案的有效性。Urban OS采用本研究项目开发的图形分析系统，并向预测和控制中心提供反馈，以优化许多社会系统和服务。我们简要介绍了我们正在进行的实现城市OS的研究项目。
In this paper, we present our ongoing research project. The objective of many ongoing research projects in high-performance computing (HPC) areas is to develop an advanced computing and optimization infrastructure for extremely large-scale graphs on the peta-scale supercomputers. The extremely large-scale graphs that have recently emerged in various application fields, such as transportation, social networks, cybersecurity, disaster prevention, and bioinformatics, require fast and scalable analysis. The Graph500 benchmark measures the performance of any supercomputer performing a breadth-first search (BFS) in terms of traversed edges per second (TEPS). In 2014–2017, our project team has achieved about 38.6TeraTEPS on K computer and been a winner at the 8th and 10th to 15th Graph500 benchmark. We commenced our research project for developing the Urban OS (Operating System) for a large-scale city in 2013. The Urban OS, which is regarded as one of the emerging applications of the cyber-physical system (CPS), gathers big data sets of the distribution of people and transportation movements by utilizing sensor technologies and storing them in the cloud storage system. In the next step, we apply optimization, simulation, and deep learning techniques to solve them and check the validity of solutions obtained on the cyberspace. The Urban OS employs the graph analysis system developed by this research project and provides a feedback to a predicting and controlling center to optimize many social systems and services. We briefly explain our ongoing research project for realizing the Urban OS.