Modified on by DinoQuintero
The residency team is pleased to announce the draft of this publication. The focus of this book is to provide you with architectural and technical details on the IBM high-performance computing (HPC) solution with the IBM AC922. There are additional components of the solution described throughout the publication. All of them working together in this clustered environment, deliver a powerful supercomputing solution to help you enable discoveries across multiple industries.
This books contains the following topics:
Part 1. Planning
Chapter 1. Introduction to IBM high-performance computing
Chapter 2. IBM Power System AC922 for high-performance computing overview
Chapter 3. Software stack
Chapter 4. Reference architecture
Part 2. Deployment
Chapter 5. Nodes and software deployment
Chapter 6. Cluster Administration and Storage Tools (CAST)
Part 3. Application development
Chapter 7. Compilation, execution, and application development
Chapter 8. Running parallel software, performance enhancement, and scalability testing
Chapter 9. Measuring and tuning applications
Appendix A. Additional material
The publication provides among others a chapter to understand the reference architecture of the solution from small to medium to large clustered environments. The idea is to describe to you that there are different supercomputing solution sizes that can help you with your challenging problems.
The overall idea of the publication is to share with you that the IBM supercomputing solution powered by the IBM AC922 can provide you with the computing capacity, capability, and power for your discoveries.
The PDF file of the book can be downloaded at the following website:
For information on IBM infrastructure solutions for high-performance computing including AI, refer to the following website:
For additional information on IBM Redbooks publications including upcoming residencies, refer to the following website:
Happy reading and success on your discoveries.
Miguel Gomez Gonzalez
Ahmad Y Hussein
Good morning IBM Redbooks readers and followers. I hope all your endeavors are going well, and I wish you success closing the first quarter of 2018.
In 2018, I am looking forward to delivering an updated technical publication on IBM high-performance computing leveraging the IBM Redbooks residency program. Over the last five years, my teams have delivered information about IBM high-performance computing solutions by way of the following publications:
High-performance computing has become a synonym solution to many clients, and lately many commercial or technical computing client have implemented some sort of high-performance computing cluster in their data center as they look to bring in clusters with faster processors (CPU) and combining them with fast GPUs to offload applications and improve time to market with their products.
IBM high-performance computing solutions combines the best of IBM hardware, software, file system, tools, open source components, networking, and others to bring to you a robust solution to help you solve challenging business problems.
I encourage you to look at the upcoming Implementing an IBM High Performance Computing (HPC) Cluster Solution on IBM Power System AC922 residency (http://w3.itso.ibm.com/residencies/PW-8401-R01) and submit your nomination. Your participation is important to the success of IBM Redbooks. Moreover, if you cannot submit your nomination, send me feedback on what you want to see within the final publication.
If you have not yet signed up to receive the IBM Redbooks newsletter with information on the latest publications, blogs, videos, and more, see the following website: https://ibm.co/2pmD9Sb.
For information about the latest IBM Redbooks publications, residencies, and workshops refer to the following website: www.ibm.com/redbooks.
Have a wonderful day, great weekend, and happy reading.
IBM Redbooks Enterprise Solutions Project Leader
During this event hosted in Prague, Czech Republic from November 6 through November 10, 2017 John Wright and Luis Bolinches delivered an Implementing High Availability and Disaster Recovery Solutions with SAP HANA on IBM Power Systems poster and presentation sessions. Traffic was continuous during the poster session, and attendees stopped by to talk to Luis and John about their experiences implementing SAP HANA on Power Systems solution. Also a presentation was delivered too which attracted 40+ participants to the session. Thank you very much to Luis and John for delivering these presentations, and to all the attendees for joining them during both presentations.
Note: Minor updates has been applied to the Implementing High Availability and Disaster Recovery Solutions with SAP HANA on IBM Power Systems, REDP-5443.
The updated publication can be downloaded at the following website:
For more information about SAP HANA on IBM Power Systems, refer to the IBM Power Systems for SAP HANA at the following website:
Rodrigo Ceron Ferreira de Castro
Los residentes del project les comunican la publicacion final del IBM Redpaper Implementing IBM Spectrum Scale, REDP-5254. Esta publicacion provee detalles de las ultimas caracteristicas anunciadas con IBM Spectrum Scale v4.1.1.
La publicacion contiene los siguientes capitulos:
- Introduction to IBM Spectrum Scale 4.1.1
- IBM Spectrum Scale implementation
- Case scenario: Protocol node
- Case scenario: Spectrum Scale Active File Management (AFM) - based disaster recovery
- Appendix A: New commands
- Appendix B: Performance monitoring
La publication puede ser encontrada en la siguiente direccion: http://www.redbooks.ibm.com/Redbooks.nsf/RedbookAbstracts/redp5254.html?Open
Su opinion es importante. Por favor envien sus comentarios por medio de este blog.
Para mas informacion acerca de Spectrum Scale, busque en la siguiente direccion:
In response to: 5 things you should know about IBM Spectrum Scale (formerly GPFS)
As far as I know, there are no plans for ARM support. However, I
would recommend to send feedback to the Spectrum Scale team with
the requirements so these can be evaluated.
Modified on by DinoQuintero
The residency team is pleased to announce the release of the IBM Redbooks publication: IBM Spectrum Scale (formerly GPFS), SG24-8254. The topics described in this publication are as follows:
- Infrastructure planning and considerations
- Management and maintenance
- Information lifecycle management (ILM)
- Active File Management (AFM)
- Backup and disaster recovery using IBM Spectrum Scale
- Problem determination
- Recovery of the IBM Spectrum Scale file system configuration
- How to obtain an IBM Spectrum Scale trial version
Note: This edition applies to IBM Spectrum Scale (formerly GPFS) v22.214.171.124 (TL1), SLES 11 SP2, SLES11 SP3, RHEL ppc64 6.5, AIX 7.1 TL3, Tivoli Storage Manager 7.1 Server, Client and Storage Agent (LAN-free), Tivoli Storage Manager for Space Management 7.1.1, and TSM for Space Management (HSM) 7.1.1.
This publication can be downloaded at the following website:
Note: This book has not been updated. This book has been placed online once again as it contains useful and still applicable concepts.
Although this publication is not at the latest Spectrum Scale releases, the book describes concepts which are still applicable to later releases of the solution. However, for the latest information about the the solution, including its latest releases, visit the IBM Knowledge Center for IBM Spectrum Scale at the following website:
Note: The initial version of this publication was released on February 23, 2015 to support the IBM re-branding announcement of the solution.
The residency team encourages the readers to send us feedback.
For additional information about IBM Spectrum Scale, refer to the following website:
5 things you should know about IBM Spectrum Scale (formerly GPFS)
Spectrum Scale now part of IBM Spectrum Storage family, part of IBM's $1 Billion investment in software defined storage.
Single global namespace across all data tiers, including flash and cloud (service on Softlayer).
Object storage with integration with Swift (will be better with 4.1.1 release, not sure you can say that in a blog).
Scalability to handle the massive, ever-growing amounts of data in the new CAMS world.
Open and flexible, available as software (on almost any hardware), as a bundled IBM solution, or as a service running on Softlayer.
This project will update the previous GPFS publication: Implementing the IBM General Parallel File System (GPFS) in a Cross Platform Environment, SG24-7844. Additional areas to explore and document are:
• GPFS as a part of IBM’s broader software-defined storage environments
• Review use cases and expand the breadth of applicability of GPFS beyond traditional HPC use cases – SAP HANA, DB2 BLU Acceleration, Hadoop Analytics, Object store for OpenStack, and so on.
• Document GPFS FPO as a part of the GPFS capability for Big Data (not covered in prior IBM RedBooks)
• Position the IBM Platform Computing cloud service for clients needing high-performance software-defined storage in the cloud
• Explain the synergy between GPFS and storage intensive IBM products: Platform Symphony, Platform LSF, Algo, DataStage (where GPFS is part of the big data references architecture), Optim, BigInsights, Streams, SPSS, and so on.
In addition to updating the GPFS publication, the residency team will explore and document the following topics:
o An overview of the Platform Computing Cloud Services
o MultiCluster technology is central to the cloud service since this is how hybrid clusters are deployed
o GPFS as an option in the cloud service and why it is valuable
o Comparative economics of deploying on-premise vs. in the cloud
For additional information including the link to the nomination form, refer to the following website:
5 Things you should know about the IBM Software Defined Environment (SDE)
1. SDE helps enable your organization to deliver IT services in the most efficient and productive way possible.
2. SDE combines compute, network, management, and storage resources and optimizes these resources efficiently.
3. SDE is the next generation of agile and optimized information technology.
4. SDE embraces open industry standards to foster freedom of choice and to help protect investments.
5. SDE leverages IBM Platform Resource Scheduler derived from IBM Platform Computing for dynamic resource management.
For additional information about the IBM Software Defined Environment, and to learn more about it, refer to the following website:
I recently announced the opening for nominations for the IBM Software Defined Environment residency program. I encourage you to submit your nomination utilizing the link included at the bottom of this announcement.
A Software Defined Environment (SDE) optimizes the entire computing infrastructure — compute, storage and network resources — so that it can adapt to the type of work required. In today's environment, resources are assigned manually to workloads; that happens automatically in a SDE. In a SDE, workloads are dynamically assigned to IT resources based on application characteristics, best-available resources, and service level policies to deliver continuous, dynamic optimization and reconfiguration to address infrastructure issues. Underlying all of this are policy-based compliance checks and updates, in a centrally managed environment. So, think integration, automation and optimization. Those are enablers to cloud delivery and analytics. SDE has the capability to accelerate business success by making a happy marriage between workloads and resources so you have a responsive, adaptive environment.
With IBM’s Software Defined Environment, infrastructure is fully programmable to rapidly deploy workloads on optimal resources and to instantly respond to changing business demands
Software = Abstracted and virtualized IT infrastructure resources managed by software
Defined = Applications automatically define infrastructure requirements and configuration
Environments = IT infrastructure that extends multiple environments to go beyond the data center
During the residency the team will position all the IBM Software Defined Environment (SDE) offerings in a holistic way, and focus on SDE planning, implementation and management. When possible, the SDE reference architecture can be describe including software products such as the IBMSmartCloud Orchestrator, Provisioning and also Application LifeCycle Management, and so on. The overall goal of the residency is to create a publication describing how to make all the bits work together with references to the content that explains how the individual products work within the SDE solution.
To read more about the residency objective, and to also submit your nomination, please refer to the following website:
If you are involved in this area, please send me your suggestions for this residency including comments about the what you think makes this solutions a viable option for your enterprise.
For additional information about the IBM Software Defined Environment, and to learn more about it, refer to the following website:
You are invited to attend on Tuesday May 20 at 10:30 AM during the IBM Edge event, session sBD06 IBM Big Data Analytics Over a Public Cloud By: Luis Cruz and Daniel de Souza Casali and discover how to implement a big data analytic environment over SoftLayer running Hadoop with GPFS and performing map reduce with IBM Symphony. You can use the information presented during this session as a reference architecture that you can run over a public cloud for your company.
For additional information on all sessions to be presented during the event, refer to:
For information about the IBM Edge event, refer to the following website:
For information on the IBM Redbooks publication which can provide more details on the reference architecture presented during the session, refer to the IBM Platform Computing Solutions Reference Architectures and Best Practices, SG24-8169 at the following website:
IBM GPFS provides fast data access and simple, cost effective data management. IBM GPFS can help:
Remove data-related bottlenecks
Simplify data management at scale
Empower global collaboration
Manage the full data lifecycle cost effectively
Ensure end-to-end data availability, reliability, and integrity
IBM GPFS - Are you Ready?
In this short video presentation at https://www.youtube.com/watch?v=y5ELxaZpo4Q
, Scott Fadden provides a brief overview of IBM GPFS. The GPFS parallel file system is currently being utilized across multiple industries worldwide to manage the growth of data. For more Information on the IBM GPFS, refer to the website:
Also, visit the IBM GPFS WIKI which contains the latest information regarding the solution, implementation, and technical information. The WIKI can be accessed at:
5 Things You Should Know About IBM HPC Central
1. Delivers information on how to start to learn about IBM High Performance Computing (HPC)
2. Contains technical documentation on IBM High Performance Computing (HPC) hardware and software solutions
3. Provides information on best practices in particular on Cluster administration best practices
4. Direct channel to provide feedback, and receive feedback on technical computing topics
5. Focuses on IBM High Performance Computing (HPC) announcements, how-to details, and special information for anyone involved with IBM Technical Computing
The High Performance Computing (HPC) Central is a joint IBM/Customer accessible and editable forum to provide improved HPC technical communications. To access this site, refer to the following website:
Announcing the IBM PowerHA SystemMirror for AIX 7.1.3
The IBM PowerHA SystemMirror for AIX Standard Edition is the IBM Power Systems data center solution that helps protect critical business applications from outages, planned or unplanned. PowerHA SystemMirror for AIX Enterprise Edition includes the Standard Edition plus advanced capabilities such as failover to back up resources at remote locations.
PowerHA SystemMirror for AIX 7.1.3 enhances and adds new capabilities, including the following key features:
1. HyperSwap for active-active two-site configurations
2. HyperSwap for single-node clusters
3. Support for clustering with unicast communications
4. Capability to dynamically modify the host name of a clustered node
5. Operator-managed manual failover policy for multisite linked clusters
6. Graphical Cluster Simulator
7. Enhanced and flexible high availability management support for SAP