Book description
This IBM® Redbooks® publication documents and addresses topics to set up a complete infrastructure environment and tune the applications to use an IBM POWER9™ hardware architecture with the technical computing software stack.
This publication is driven by a CORAL project solution. It explores, tests, and documents how to implement an IBM High-Performance Computing (HPC) solution on a POWER9 processor-based system by using IBM technical innovations to help solve challenging scientific, technical, and business problems.
This book documents the HPC clustering solution with InfiniBand on IBM Power Systems™ AC922 8335-GTH and 8335-GTX servers with NVIDIA Tesla V100 SXM2 graphics processing units (GPUs) with NVLink, software components, and the IBM Spectrum™ Scale parallel file system.
This solution includes recommendations about the components that are used to provide a cohesive clustering environment that includes job scheduling, parallel application tools, scalable file systems, administration tools, and a high-speed interconnect.
This book is divided into three parts: Part 1 focuses on the planners of the solution, Part 2 focuses on the administrators, and Part 3 focuses on the developers.
This book targets technical professionals (consultants, technical support staff, IT architects, and IT specialists) who are responsible for delivering cost-effective HPC solutions that help uncover insights among clients' data so that they can act to optimize business results, product development, and scientific discoveries.
Table of contents
- Front cover
- Notices
- Preface
- Part 1 Planning
- Chapter 1. Introduction to IBM high-performance computing
- Chapter 2. IBM Power System AC922 server for HPC overview
- Chapter 3. Software stack
- Chapter 4. Reference architecture
- Part 2 Deployment
-
Chapter 5. Nodes and software deployment
- 5.1 Deployment overview
- 5.2 System management
-
5.3 xCAT deployment overview
- 5.3.1 xCAT database: Objects and tables
- 5.3.2 xCAT node booting
- 5.3.3 xCAT node discovery
- 5.3.4 xCAT baseboard management controller discovery
- 5.3.5 xCAT installation types: Disks and state
- 5.3.6 xCAT network interfaces: Primary and additional
- 5.3.7 xCAT software kits
- 5.3.8 xCAT synchronizing files
- 5.3.9 xCAT version
- 5.3.10 xCAT scenario for high-performance computing
- 5.4 Initial xCAT management node installation on an IBM Power System LC922 server
-
5.5 xCAT node discovery
- 5.5.1 Verification of network boot configuration and Genesis image files
- 5.5.2 Configuring the DHCP dynamic range
- 5.5.3 Configuring BMCs to DHCP mode
- 5.5.4 Definition of temporary BMC objects
- 5.5.5 Defining node objects
- 5.5.6 Configuring the host table, DNS, and DHCP servers
- 5.5.7 Booting into node discovery
-
5.6 xCAT compute nodes (stateless)
- 5.6.1 Network interfaces
- 5.6.2 Red Hat Enterprise Linux operating system images
- 5.6.3 NVIDIA CUDA Toolkit
- 5.6.4 Mellanox OpenFabrics Enterprise Distribution
- 5.6.5 IBM XL C/C++ runtime libraries
- 5.6.6 IBM XL Fortran runtime libraries
- 5.6.7 Advance Toolchain runtime libraries
- 5.6.8 IBM Spectrum MPI
- 5.6.9 IBM Parallel Performance Toolkit
- 5.6.10 IBM Engineering and Scientific Subroutine Library
- 5.6.11 IBM Parallel Engineering and Scientific Subroutine Library
- 5.6.12 IBM Spectrum Scale (formerly IBM GPFS)
- 5.6.13 PGI runtime libraries
- 5.6.14 IBM Spectrum LSF integration with Cluster Systems Management
- 5.6.15 Synchronizing the configuration files
- 5.6.16 Generating and packing the image
- 5.6.17 Node provisioning
- 5.6.18 Postinstallation verification
- 5.7 xCAT login nodes (stateful)
-
Chapter 6. Cluster Administration and Storage Tools
- 6.1 Cluster Systems Management
-
6.2 Preparing CSM
- 6.2.1 Software dependencies
- 6.2.2 Installation
- 6.2.3 CSM RPMs overview
- 6.2.4 Installing CSM on to the management node
- 6.2.5 Installing CSM on to the service node
- 6.2.6 Installing CSM in the login, launch, and workload manager nodes
- 6.2.7 Installing CSM in the compute nodes
- 6.2.8 Configuration
- 6.2.9 Configuring the CSM database
- 6.2.10 Default configuration files
- 6.2.11 Configuring SSL
- 6.2.12 Heartbeat interval
- 6.2.13 Environmental buckets
- 6.2.14 Prolog and epilog scripts
- 6.2.15 CSM Pluggable Authentication Module
- 6.2.16 Starting the CSM daemons
- 6.2.17 Running the infrastructure health check
- 6.2.18 Setting up the environment for job launch
- 6.2.19 Installing the configuring the CSM REST daemon
- 6.2.20 Uninstalling the CSM daemons
- 6.2.21 Diskless images
- 6.3 Burst Buffer
- Part 3 Application development
-
Chapter 7. Compilation, execution, and application development
- 7.1 Compiler options
- 7.2 Porting applications to IBM Power Systems servers
- 7.3 IBM Engineering and Scientific Subroutine Library
- 7.4 IBM Parallel Engineering and Scientific Subroutine Library
- 7.5 Using POWER9 vectorization
-
7.6 Development models
- 7.6.1 OpenMP programs with IBM Parallel Environment
- 7.6.2 CUDA C programs with the NVIDIA CUDA Toolkit
- 7.6.3 OpenACC
- 7.6.4 IBM XL C/C++ and Fortran offloading
- 7.6.5 MPI programs with IBM Parallel Environment V2.3
- 7.6.6 Hybrid MPI and CUDA programs with IBM Parallel Environment
- 7.6.7 OpenSHMEM programs in IBM Parallel Environment
- 7.6.8 Parallel Active Messaging Interface programs
- 7.6.9 MPI programs that use IBM Spectrum MPI
- 7.6.10 Migrating from an IBM Parallel Environment Runtime Edition environment to IBM Spectrum MPI
- 7.6.11 Using IBM Spectrum MPI
- Chapter 8. Running parallel software, performance enhancement, and scalability testing
-
Chapter 9. Measuring and tuning applications
- 9.1 Effects of basic performance tuning techniques
-
9.2 General methodology of performance benchmarking
- 9.2.1 Defining the purpose of performance benchmarking
- 9.2.2 Benchmarking plans
- 9.2.3 Defining the performance metric and constraints
- 9.2.4 Defining the success criteria
- 9.2.5 Correctness and determinacy
- 9.2.6 Keeping the log of benchmarking
- 9.2.7 Probing scalability
- 9.2.8 Evaluation of performance on a favorable number of cores
- 9.2.9 Evaluation of scalability
- 9.2.10 Conclusions
- 9.2.11 Summary
- 9.3 Sample code for the construction of thread affinity strings
- 9.4 IBM Engineering and Scientific Subroutine Library performance results
- 9.5 GPU tuning
- 9.6 Application development and tuning tools
- Appendix A. Additional material
- Related publications
- Back cover
Product information
- Title: IBM High-Performance Computing Insights with IBM Power System AC922 Clustered Solution
- Author(s):
- Release date: May 2019
- Publisher(s): IBM Redbooks
- ISBN: 9780738457451
You might also like
book
Implementing an IBM High-Performance Computing Solution on IBM POWER8
This IBM® Redbooks® publication documents and addresses topics to provide step-by-step programming concepts to tune the …
book
IBM Platform Computing Integration Solutions
This IBM® Redbooks® publication describes the integration of IBM Platform Symphony® with IBM BigInsights™. It includes …
book
Highly Efficient Data Access with RoCE on IBM Elastic Storage Systems and IBM Spectrum Scale
With Remote Direct Memory Access (RDMA), you can make a subset of a host's memory directly …
book
Using HPC for Computational Fluid Dynamics
Using HPC for Computational Fluid Dynamics: A Guide to High Performance Computing for CFD Engineers offers …