This post is password protected. To view it please enter your password below:

Password:

10 Years adding value to IBM Power Systems

AUCKLAND, NZ – October 26, 2011 – Spectrum Consulting, a professional services company that provides value-add solutions around IBM Power Systems and Storage is turning 10 next month. Spectrum started from humble beginnings on the 7th November 2001 and by focusing on providing turn-key solutions on IBM Power Systems has seen year on year growth. Spectrum partners very closely with IBM which enables us to add more value around IBM Power Systems infrastructure from systems design / architecture through to implementation and migrations (including Disaster Recovery)

“Our business has really evolved with the technology over the years and the opportunity for customers to benefit from IBM Power Systems has never been stronger” said managing directer and founder of Spectrum Consulting -Paul Tomlinson. “We started in 2001 with the mission to be the best IBM UNIX solutions provider in New Zealand and this goal is becoming clearer as we win more and more customers who wish to get the best out of their IBM Power Systems”

Spectrum Consulting credits its contined growth to a strong customer base, excellent value-add services offering using the highest level of skills in New Zealand. Spectrums focus has enabled it to become the boutique vendor of choice when customers are looking for IBM Power Systems solutions.

The following recognition will be made at the Awards Evening on 10 March 2011:

• Power Business Partner of the Year
• System Storage Business Partner of the Year
• System x Business Partner of the Year
• Software Business Partner of the Year

Spectrum was the winner of the:
2010 Power Business Partner of the Year

Six resellers were given awards in recognition of “revenue attainment, ease of doing business with IBM, year-on-year growth in IBM business, participation in the channel programs, enablement, co-marketing and innovation,” according to Raymond Skoglund, IBM’s New Zealand channel manager.

IBM BP Awards logo

Auckland, NZ – Mar 4th, 2011 – Spectrum Consulting, New Zealand premier provider of IBM Power Systems sales and support was announced today as a finalist in the Dynamic Infrastructure category of the 2010 IBM Business Partner Awards for its storage solution at Weta Productions in Wellington, New Zealand.

“Congratulations to all of the IBM Business Partner finalists,” says Raymond Skoglund, IBM New Zealand’s channel manager. “This year’s finalists have all demonstrated an ability to readily identify the business requirements of their customers and then deliver solutions that add significant value. This year we have seen customers develop new revenue streams, enhance visibility into their planning processes and increase performance while reducing costs and mitigating risk. The high quality of all of the submissions have made the selection difficult indeed and we are extremely pleased to have such talented people and innovative organizations as our IBM Business Partners.”

What is Dynamic Infrastructure?
A Dynamic Infrastructure is designed for today’s instrumented world, helping clients integrate the growing intelligent business infrastructure with the necessary underlying design of a flexible, secure and seamlessly managed IT infrastructure.

A Dynamic Infrastructure:

• Helps reduce operational cost and complexity, and achieve breakthrough productivity gains through virtualisation, optimisation and energy efficiency.
• Addresses today’s security, business continuity and compliance challenges, and helps prepare for the new risks posed by a connected and collaborative world
• Allows companies to quickly respond to customers’ needs today and add new applications and technologies for future growth.

Organisations supported by a dynamic infrastructure can deliver superior business and IT services with agility and speed.

PROJECT: Weta Workshop has massive requirements for storage as they ‘render’ the millions and millions of images that go into creating award-winning animated television shows. But their infrastructure couldn’t scale fast or far enough to support their anticipated High Definition workloads. Spectrum Consulting specified and assisted in the installation of the IBM Scale Out Network Attached Storage (SONAS) solution, one of the first such implementations worldwide. Weta Workshop can now produce significantly more rendered sequences than before and now has the capacity to take on additional projects.

About the IBM Business Partner Awards

The IBM NZ Business Partner Awards, in association with Reseller News, recognise and reward the achievements of IBM Business Partners and IT resellers who are leading the way by helping to deliver smarter solutions, whether through outstanding industry knowledge, superior service or timely delivery of IT solutions. The awards are for Business Partners who are leading the charge and have delivered unique and innovative solutions that have raised the standard for business excellence and customer satisfaction. Winners will be announced in Auckland on 10 March, and will be published in the Reseller News 18 March 2011 edition. For more information please visit www.bpawards.co.nz.

This technical white paper demonstrates the extent of the performance lead that PowerVM enjoys over x86-based add-on virtualization products, by running identical virtualized workload benchmarks on comparable POWER7- and Intel-based systems. The benchmark results show the dramatic improvements that can be expected when deploying virtualized workloads on Power Systems servers, compared to the performance of those same workloads on x86-based platforms running using a third-party virtualization product such as VMware vSphere 4. Key findings include the following:

* PowerVM on Power 750 performs up to 65% better than VMware
* PowerVM on Power 750 scales to four times more virtual CPUs than VMware in a virtual machine
* PowerVM on Power 750 scales linearly to use all CPUs, while VMware does not

It is clear from the published benchmark results that PowerVM on POWER7-based platforms not only offers vastly superior scalability than VMware vSphere on Intel x86-based servers, but it also makes more efficient use of system resources and imposes a negligible impact on performance. Many of these advantages relate to the fact that PowerVM is built directly into the firmware of all Power Systems servers, as opposed to x86-based virtualization products such as VMware vSphere, which are typically third-party software add-ons that are sold and installed separately.

In summary, the benchmark results published in this white paper prove that PowerVM on POWER7 platforms offers a far superior virtualization solution than VMware vSphere on Intel x86 platforms, with higher performance, broader scalability and increased flexibility.

View full report ‘A Comparison of PowerVM and VMware Virtualization Performance’ (175KB)

IBM PowerVM provides industrial-strength virtualization for Power Systems™ servers running AIX, IBM i, and Linux operating systems. PowerVM now features Active Memory Sharing, the technology that allows you to intelligently exchange memory between running partitions for increased optimization of physical memory resources. Active Memory Sharing enables the sharing of a pool of physical memory among logical partitions (LPARs) on a single server, helping to increase memory utilization and drive down system costs.

This paper provides a high-level overview of Active Memory Sharing including an overview of its capabilities, a description of its benefits, and examples of good candidate workloads as well as usage considerations.

Active Memory Sharing Whitepaper

Overview of shared memory

Shared memory is physical memory that is assigned to the shared memory pool and shared among multiple logical partitions. The shared memory pool is a defined collection of physical memory blocks that are managed as a single memory pool by the hypervisor. Logical partitions that you configure to use shared memory (hereafter referred to as shared memory partitions) share the memory in the pool with other shared memory partitions.

For example, you create a shared memory pool with 16 GB of physical memory. You then create three logical partitions, configure them to use shared memory, and activate the shared memory partitions. Each shared memory partition can use the 16 GB that are in the shared memory pool.

The hypervisor determines the amount of memory allocated from the shared memory pool to each shared memory partition based on the workload and memory configuration of each shared memory partition. When allocating the physical memory to the shared memory partitions, the hypervisor ensures that each shared memory partition can access only the memory allocated to the shared memory partition at any given time. A shared memory partition cannot access the physical memory allocated to another shared memory partition.

The amount of memory that you assign to the shared memory partitions can be greater than the amount of memory in the shared memory pool. For example, you can assign 12 GB to shared memory partition 1, 8 GB to shared memory partition 2, and 4 GB to shared memory partition 3. Together, the shared memory partitions use 24 GB of memory, but the shared memory pool has only 16 GB of memory. In this situation, the memory configuration is considered overcommitted.

Overcommitted memory configurations are possible because the hypervisor virtualizes and manages all of the memory for the shared memory partitions in the shared memory pool as follows:

1. When shared memory partitions are not actively using their memory pages, the hypervisor allocates those unused memory pages to shared memory partitions that currently need them. When the sum of the physical memory currently used by the shared memory partitions is less than or equal to the amount of memory in the shared memory pool, the memory configuration is logically overcommitted. In a logically overcommitted memory configuration, the shared memory pool has enough physical memory to contain the memory used by all shared memory partitions at one point in time. The hypervisor does not need to store any data in auxiliary storage.

2. When a shared memory partition requires more memory than the hypervisor can provide to it by allocating unused portions of the shared memory pool, the hypervisor stores some of the memory that belongs to a shared memory partition in the shared memory pool and stores the remainder of the memory that belongs to the shared memory partition in auxiliary storage. When the sum of the physical memory currently used by the shared memory partitions is greater than the amount of memory in the shared memory pool, the memory configuration is physically overcommitted. In a physically overcommitted memory configuration, the shared memory pool does not have enough physical memory to contain the memory used by all the shared memory partitions at one point in time. The hypervisor stores the difference in auxiliary storage. When the operating system attempts to access the data, the hypervisor might need to retrieve the data from auxiliary storage before the operating system can access it.

Because the memory that you assign to a shared memory partition might not always reside in the shared memory pool, the memory that you assign to a shared memory partition is logical memory. Logical memory is the address space, assigned to a logical partition, that the operating system perceives as its main storage. For a shared memory partition, a subset of the logical memory is backed up by physical main storage (or physical memory from the shared memory pool) and the remaining logical memory is kept in auxiliary storage.

A Virtual I/O Server logical partition provides access to the auxiliary storage, or paging space devices, required for shared memory partitions in an overcommitted memory configuration. A paging space device is a physical or logical device that is used by a Virtual I/O Server to provide the paging space for a shared memory partition. The paging space is an area of nonvolatile storage used to hold portions of a shared memory partition’s logical memory that do not reside in the shared memory pool. When the operating system that runs in a shared memory partition attempts to access data, and the data is located in the paging space device that is assigned to the shared memory partition, the hypervisor sends a request to a Virtual I/O Server to retrieve the data and write it to the shared memory pool so that the operating system can access it.

On systems that are managed by a Hardware Management Console (HMC), you can assign up to two Virtual I/O Server (VIOS) logical partitions to the shared memory pool at a time (hereafter referred to as paging VIOS partitions). When you assign two paging VIOS partitions to the shared memory pool, you can configure the paging space devices such that both paging VIOS partitions have access to the same paging space devices. When one paging VIOS partition becomes unavailable, the hypervisor sends a request to the other paging VIOS partition to retrieve the data on the paging space device.

You cannot configure paging VIOS partitions to use shared memory. Paging VIOS partitions do not use the memory in the shared memory pool. You assign paging VIOS partitions to the shared memory pool so that they can provide access to the paging space devices for the shared memory partitions that are assigned to the shared memory pool.

Driven by workload demands from the shared memory partitions, the hypervisor manages overcommitted memory configurations by continually performing the following tasks:

* Allocating portions of physical memory from the shared memory pool to the shared memory partitions as needed
* Requesting a paging VIOS partition to read and write data between the shared memory pool and the paging space devices as needed

The ability to share memory among multiple logical partitions is known as the PowerVM™ Active Memory Sharing technology. The PowerVM Active Memory Sharing technology is available with the PowerVM Enterprise Edition for which you must obtain and enter a PowerVM Editions activation code.

Most of IBM courses include plenty of hands-on lab exercises to reinforce the lectures and let you practice the skills taught in class. You will leave class with the added confidence that only comes from hands-on experience. IBM is the right source for all of your System p, AIX and UNIX training.

The 2011 Training schedule… IBMSchedules_Systemp

Contact Specturm today for a 10% discount off all IBM Training…

This IBM® Redbooks® publication focuses on the enhancements to IBM AIX® Version 7.1 Standard Edition. It is intended to help system administrators, developers, and users understand these enhancements and evaluate potential benefits in their own environments.

AIX Version 7.1 introduces many new features, including:
- Domain Role Based Access Control
- Workload Partition enhancements
- Topas performance tool enhancements
- Terabyte segment support
- Cluster Aware AIX functionality

AIX Version 7.1 offers many other new enhancements, and you can explore them all in this publication.
For clients who are not familiar with the enhancements of AIX through Version 5.3, a companion publication, AIX Version 6.1 Differences Guide, SG24-7559, is available.

Chapter 1. Application development and debugging
Chapter 2. File systems and storage
Chapter 3. Workload Partitions and resource management
Chapter 4. Continuous availability
Chapter 5. System management
Chapter 6. Performance management
Chapter 7. Networking
Chapter 8. Security, autentication, and authorization
Chapter 9. Installation, backup, and recovery
Chapter 10. National language support
Chapter 11. Hardware and graphics support

Document can be downloaded here

SONAS is a multi-petabyte scale out NAS storage offering for information storage and distribution. Designed to scale out to store millions to billions of active files in a single namespace. It combines massive scalability with automated data management to efficiently deliver information around the world.
Watch this space for more information!

The technical team at Spectrum Consulting recently finished a Power Systems Audit of a large publicly listed company in NZ. The audit looked at the Power Systems Hardware, Virtualization, AIX Logical Partitions, and the attached IBM SAN / Storage Servers. The environment was absolutely mission-critical to the business, running the core business applications.

The customer didn’t have the skills to implement the SAP on IBM Power Systems solution. They had simply trusted a NZ ‘Tier 1 Reseller’ to implement the solution correctly. After the implementation the customer was not provided with any documentation and therefore had a poor understanding of their environment simply believing it to be setup correct.

During the audit Spectrum discovered three separate configuration errors (Single Points of Failure) in the implementation which could have resulted in major disruption and huge losses for the entire company.

1.) The Virtual I/O server which provided the Operating System disks for the production LPARS had not been mirrored. This meant that the failure of a single hard drive would have taken 4 production systems offline.

2.) The main production LPAR had itself not been mirrored. Again the failure of a single drive would have resulted in the customer’s production system going down. Given this LPAR’s storage was being provided by the un-mirrored Virtual I/O server the risk was doubled!

3.) A production LPAR was mirrored with both copies of the mirror on the same physical hard drive. This poor implementation again meant that the loss of a single hard drive or SCSI adapter would have resulted in the failure of the LPAR.

These systems were implemented almost three years ago and the customer was totally shocked to find their production systems not configured correctly. The customer is grateful that Spectrum undertook the system audit, immediately notifying them of the issues and fixing them promptly.

The customer has engaged with Spectrum to support their IBM AIX / Power Systems environment moving forward…

If you are responsible for IBM Power Systems and would like Spectrum to audit your environment please call us today… +64 9 826 5588

Paul Tomlinson
Managing Director