- Browse
- » The student supercomputer challenge guide: from supercomputing competition to the next HPC generation
The student supercomputer challenge guide: from supercomputing competition to the next HPC generation
Author
Publisher
Science Press
Publication Date
[2018]
Language
English
Description
Loading Description...
Table of Contents
From the Book
Machine generated contents note: 1.Development and Application of Supercomputing
1.1.Why Do We Study Supercomputing?
1.2.Development and Architecture Classification of Supercomputers
1.2.1.Development of Supercomputers
1.2.2.Architecture Classification of Supercomputers
1.2.3.The Development Trend of Supercomputers
1.2.4.The Rise and Development of Heterogeneous Architectures
1.3.Current Status and Challenges of Supercomputing Applications
1.3.1.Current Status of Supercomputing Applications
1.3.2.Challenges of Supercomputing Applications
2.Construction and Power Management of Supercomputing System
2.1.Components of a Supercomputing System
2.2.Power Monitoring and Management of a Supercomputing System
2.3.Building a Performance-Balanced HPC System
2.3.1.A Performance-Balanced HPC System
2.3.2.How to Manage the HPC System
2.3.3.Monitoring Management
2.3.4.Job Scheduling
2.3.5.The 10-Teraflops HPC System Design
3.Network Communication in a Supercomputing System
3.1.Overview of the InfiniBand Technology
3.1.1.Host Channel Adapter (HCA)
3.1.2.Target Channel Adapter (TCA)
3.1.3.Switch
3.1.4.Router
3.1.5.Cable and Connection Module
3.2.The Current Status of InfiniBand in HPC
3.3.Core Technology of InfiniBand RDMA
3.3.1.A Brief Introduction to RMDA
3.3.2.Core Technologies of RDMA
3.4.Optimization of HPC Applications Based on the InfiniBand Network
3.4.1.MPI Communication Protocol
3.4.2.MPI Function
3.4.3.Performance Comparison of MPI on Different Networks
3.4.4.Application Performance
3.4.5.MPI Application Optimization
3.5.InfiniBand Accelerates Cloud Computing and Big Data
4.Building the Environment for HPC Applications
4.1.Applied Environment for CPU Parallel System
4.1.1.Hardware Environment
4.1.2.Software Environment
4.1.3.The Development Environment
4.2.The Application Environment of CPU + MIC Heterogeneous Parallel Systems
4.2.1.The Hardware Environment
4.2.2.The Software Environment
4.2.3.The Development Environment
4.2.4.Further Reading
4.3.The Application Environment of CPU + GPU Heterogeneous Parallel Systems
4.3.1.The Hardware Environment
4.3.2.The Software Environment
4.3.3.The Development Environment
4.3.4.Further Reading
5.Supercomputer System Performance Evaluation Methods
5.1.Introduction to HPC System Performance Evaluation
5.1.1.The Significance and Challenge of System Performance Evaluation
5.1.2.The Metrics and Main Contents of System Performance Evaluation
5.1.3.The Procedures and Methods of System Performance Evaluation
5.2.HPC Systems Performance Evaluation Methods
5.2.1.The Main Evaluation Programs and Applications
5.2.2.High-Performance Computing System Rankings
5.2.3.Further Reading
5.3.HPC Application Analysis and Monitoring Tools
5.3.1.The Extraction of Application Characteristic Data
5.3.2.The Analysis of Application Characteristics Experimental Induction
5.3.3.The Analysis of Application Characteristics Theoretical Predictions
5.3.4.Application Characteristic Monitoring Tool Teye
6.An Introduction to International Student Supercomputing Competitions
6.1.The ASC Student Supercomputer Challenge
6.1.1.History
6.1.2.Participating Teams
6.2.ISC Student Cluster Challenge
6.2.1.History
6.2.2.Participating Teams
6.2.3.Patterns and Trends
6.2.4.Further Reading
6.3.SC Student Cluster Competition
6.3.1.History
6.3.2.Participating Teams
6.3.3.Patterns and Trends
6.3.4.Further Reading
7.History and Prospects of ASC Student Supercomputer Challenge
7.1.The First Chinese Student Supercomputer Challenge
7.1.1.The First International Competition in China
7.1.2.The Competition Process
7.1.3.The Competition's Three Key Elements: College Students, International Standards and Practical Skills
7.2.Asian Student Supercomputer Challenge 2013 (ASC13)
7.2.1.ASC13's Orientation and Purpose
7.2.2.The Competition Rules, Schedules and Organizational Structures
7.2.3.The Ten Finalists of ASC13
7.2.4.The Jury Committee's Comprehensive Evaluation on the Participating Universities
7.3.Student Supercomputer Challenge 2014 (ASC14)
7.3.1.Eighty-Two Participating Universities, the Largest Scale in History
7.3.2.Gathering of International Supercomputing Experts
7.3.3.Preliminary Competition: Four Tasks for Selecting the Finalists
7.3.4.Final Competition: Seven Tasks Challenging the Finalists' Skill Limit
7.3.5.Global Young Talents Challenging the Performance Limit of the World's Fastest Supercomputer
7.3.6.Dark Horse Emerging
7.3.7.ePrize Computing Challenge Award for Motivating Global Supercomputing Talent
7.4.The Asia Student Challenge 2015 (ASC15)
7.4.1.A Historical One with 152 Teams from Six Continents
7.4.2.A Gathering of Specialists from Around the World
7.4.3.ASC Preliminary: Challenge Covering Molecule to the Universe
7.4.4.ASC Final: Close to Application
7.4.5.A Gathering of Young Talent to Tackle Top Challenges
7.4.6.A Bonus of 250,000 for the Final
7.4.7.Zongci Liu "Presented" in the Closing Ceremony
8.Rules of ASC Student Supercomputer Challenge
8.1.Rules of the ASC14 Preliminary Stage
8.1.1.Brief Introduction of the Preliminary Stage
8.1.2.Detailed Rules of the Preliminary Stage
8.1.3.Summary of the Preliminary Stage
8.2.Rules of the ASC14 Final Stage
8.2.1.Introduction to the Final Stage
8.2.2.Detailed Rules of the Final Stage
8.2.3.Summary of the Final Stage
9.Competition Proposal
9.1.Introduction
9.2.Team Selection and Composition
9.3.Basic Requisite
9.4.Training
9.4.1.Training Plan
9.5.Training Hardware
9.5.1.Hardware Platform
9.6.Competition Proposal
9.6.1.Proposal Sections
9.6.2.Guidelines for Graphs and Diagrams Presentation
9.6.3.Guidelines for Other Sections
9.6.4.Knowledge of High Performance Computing Activities
9.6.5.Hardware Design and Energy Efficiency
9.6.6.High Performance LINPACK (HPL)
9.6.7.Application Software
9.6.8.Configurations and Assumptions
9.6.9.Code Optimization
9.7.Task Distribution
9.8.Summary
10.Design and Construction of the Clusters for the Competition
10.1.Architecture of Cluster
10.2.Performance and Power Consumption of Cluster Systems
11.Optimization for the High Performance LINPACK Benchmark
11.1.Brief Introduction of HPL
11.2.Installation and Test
11.2.1.Compiling ATLAS
11.2.2.Compiling HPL
11.2.3.Configuration Modification
11.2.4.HPL Testing
11.3.General Optimizations for HPL
11.3.1.Modifying the Configuration Parameters
11.3.2.Selecting the Third-Party Library
11.3.3.Compiler Optimizations
11.3.4.The Relation Between Node and Process
11.3.5.Task Allocation Optimizations
11.3.6.Heat Dissipation
11.3.7.Further Reading
12.Optimization for the Molecular Dynamics Software GROMACS
12.1.Introduction
12.2.Experimental Environment and Hardware Configuration
12.3.Parameters
12.4.Performance Analysis of the Software Library
12.4.1.Compiler
12.4.2.MPI Tool
12.4.3.Fast Fourier Transform (FFT) Library
12.4.4.Summary
12.5.Performance Analysis of Computing Resource Allocation
12.5.1.Number of Cores
12.5.2.Number of Cores on Each Node
12.5.3.Number of Compute Nodes
12.5.4.Summary
12.6.Performance Test Using GPU
12.7.Further Reading
13.Optimizations for Ocean Model LICOM
13.1.Introduction to LICOM
13.1.1.LICOM
13.1.2.The Development of LICOM
13.1.3.Necessity of LICOM Acceleration
13.2.Analysis of the Key Optimization Points of LICOM
13.3.Hardware Environment
13.4.Software Environment Configuration
13.4.1.Install and Configure the NetCDF C Library
13.4.2.Install the NetCDF Fortran library
13.4.3.LICOM Installation
13.5.Implementation and Result
13.5.1.The
Implementation
13.5.2.Optimization Results
13.6.Review of Our Preparation for ASC14
13.7.Review of Optimizing LICOM in ASC14 Final
13.8.LICOM Optimization Summary and Evaluation
13.9.Further Reading
14.Optimization of Three Dimensional Elastic Wave Modeling Software 3D_EW
14.1.Introduction to the Team and the Preparation
14.1.1.Introduction to the Team
14.1.2.Technical Preparation
14.2.Explanation of the ASC14 3D_EW Rules
14.3.Introduction to 3D_EW
14.4.Analysis of the 3D_EW Program
14.4.1.Hotspot Analysis
14.4.2.Code Analysis
14.5.Parallel Design and Implementation of 3D_EW on the Hybrid CPU + MIC Cluster
14.5.1.The General Parallel Design for 3D_EW Based on the Hybrid CPU + MIC Cluster
14.5.2.Implementation of 3D_EW Based on CPU + MIC Cluster
14.6.The Experiment and Analysis of 3D_EW Optimization Based on Tianhe-2
14.6.1.Environment of the Platform
14.6.2.Single-Node Performance
14.6.3.The Performance Based on Inter-node Platform.
Excerpt
Loading Excerpt...
Author Notes
Loading Author Notes...
More Details
ISBN
9789811037306
Staff View
Loading Staff View.

