高性能与大数据计算培训大纲 High-Performance and Data-Intensive Computing
类别：未知 发布人：admin 浏览次数： 次 发布时间：2018-07-09 17:17
时间：2018.7.9 - 2018.7.14   地点：外围投注365B404
教授级高级工程师 刘龑 （Yan Liu），美国 UIUC

教师简介 (Instructor’s Biography)

Dr. Yan Liu is Director of Technical Program at the CyberGIS Center for Advanced Digital and Spatial Studies and Senior Research Programmer (equivalent to the academic title of Professorate Senior Engineer in China) at the National Center for Supercomputing Applications and the Department of Geography and Geographic Information Science at the University of Illinois at Urbana-Champaign. He obtained his PhD in Informatics from the University of Illinois at Urbana-Champaign, MS from the University of Iowa, and BS and ME from Wuhan University. His research interests include scientific computing algorithms, software, and systems, high-performance heuristic algorithms for spatial optimization, and geospatial data science. At the CyberGIS Center, he is the lead software architect of CyberGIS science gateway and middleware. He is a staff scientist in XSEDE, a leading national cyberinfrastructure in U.S., to provide advanced scientific computing consulting. He has published over 40 peer-reviewed journal and conference papers. His work in continental-scale high-resolution flood mapping and political redistricting have received broad media coverage by the U.S. National Science Foundation (NSF), HPC Wire, Top 500, WIRED, Communications of the ACM, and Nature.

课程介绍 (Course Description)

This intermediate level training course on high-performance computing (HPC) and data-intensive computing teaches mainstream technologies and programming frameworks in cluster computing, multicore/manycore computing, scalable parallel computing, and data-intensive computing. This course selects teaching materials from public tutorials and classes that U.S. supercomputing centers offer to academic users. The teaching of related concepts, methods, tools/libraries, and systems aims to help attendees understand computational thinking in parallel computing context for solving scientific computing problems in various application domains. A series of hands-on exercises is thus designed for attendees to apply parallel computing techniques in programming assignments. After the training, attendees should feel comfortable to explore insights of identifying and leveraging parallelisms in specific problems and parallelizing sequential software code.

教学内容与计划 (Topics and Schedule)

The 6-day intensive training course covers a broad spectrum of topics. The teaching of each topic listed below includes two parts: lecturing and hands-on training. The hands-on training includes multiple programming exercises.

Registration

To participate, attendees must register for this training course before July 09. Please follow our announcement to apply and register. After your application is approved, please follow the instructions to test your laptop connection to WHU HPC and login.

Schedule

课前预备日 1（Preparation day 1, July 07）
09:00 - 17:00 Collaboration with WHU HPC to install, deploy, configure, and test HPC training software and codes as well as dependent system libraries
课前预备日 2（Preparation day 2, July 08）
09:00 - 17:00 Collaboration with WHU HPC to install, deploy, configure, and test data-intensive computing training software, services, virtual machines, and containers
第一天（Day 1, July 09）
09:00 - 09:50 Introduction to Supercomputing and Data-Intensive Computing
10:00 - 10:50 Hands-on: Accessing WHU Cluster Computing Environment
11:00 - 12:00 Resource Management and Job Submission
14:00 - 14:50 Hands-on: Cluster Computing 101
15:00 - 15:50 Embarrassingly Parallel Computing
16:00 - 17:00 Hands-on: Tools for Embarrassingly Parallel Computing
第二天（Day 2, July 10）
09:00 - 09:50 Introduction to Scientific Workflows on HPC
10:00 - 10:50 Hands-on: Native Workflow Support in Job Schedulers
11:00 - 12:00 Exascale Computing: Opportunities and Challenges
14:00 - 14:50 Manycore and Multicore Computing
15:00 - 15:50 OpenMP Programming
16:00 - 17:00 Hands-on: OpenMP Programming
第三天（Day 3, July 11）
09:00 - 09:50 Introduction to Message Passing Interface (MPI)
10:00 - 10:50 Hands-on: MPI Basics
11:00 - 12:00 MPI Point-to-point Communication
14:00 - 14:50 Hands-on: MPI Point-to-point Communication Examples
15:00 - 15:50 MPI Collective Communication
16:00 - 17:00 Hands-on: MPI Collective Communication Examples
第四天（Day 4, July 12）
09:00 - 09:50 MPI Non-blocking Communication
10:00 - 10:50 Hands-on: Blocking vs. Non-blocking Communication
11:00 - 12:00 MPI IO
14:00 - 14:50 Hands-on: MPI IO exercises
15:00 - 15:50 MPI One-sided Communication
16:00 - 17:00 Other Advanced MPI Features
第五天（Day 5, July 13）
09:00 - 09:50 Introduction to Big Data Computing
10:00 - 10:50 Hands-on: MapReduce Programming
11:00 - 12:00 Introduction to Spark
14:00 - 14:50 Hands-on: Spark Basics
15:00 - 15:50 Spark Programming
16:00 - 17:00 Hands-on: Spark for Big Data Analysis
第六天（Day 6, July 14）
09:00 - 09:50 Container Technologies
10:00 - 10:50 Hands-on: Docker Container Exercises
11:00 - 12:00 Containers on HPC
14:00 - 14:50 Hands-on: Singularity
15:00 - 15:50 Jupyter: Bridging HPC and Researchers for Reproducible and Interactive Analysis
16:00 - 17:00 Hands-on: Jupyter Environment for Interactive Computing
 
 

邀请人：蔡朝辉

上一篇：7月3日学术报告（宫明明，美国卡耐基梅隆大学）
下一篇：7月12日学术报告（Steve Cassidy 澳大利亚麦考瑞大学）