This chart deck is intended for education related to AIX in 2Q 2010
This is the Seller presentation and is not intended for use with clients
Jay Kruemcke
AIX Program Director
[email_address]
Austin, TX
<number>
All around the world, businesses, governments and institutions are investing to reduce costs, drive innovation and transform their infrastructure. The economic downturn has intensified this trend, as leaders seek not simply to repair what is broken, but to prepare for a 21st Century economy.
IBM has been talking about our vision for a smarter planet for close to a year now- and working with thousands of clients - with great success and traction taking hold. There’s no doubt that in almost any industry today, digital and physical infrastructures are coming together more and more. To some extent – today - almost anything – any person, object, process or service -- can become digitally aware and connected. Every industry is experiencing the benefits, and feeling the challenges, being presented by a smarter planet that is more instrumented, interconnected and intelligent.
For example:
Cities large and small depend on the ability of Power Systems™ to sift through the data needed to not only solve crimes and respond to emergencies, but to help prevent them. Power Systems help manage traffic, share information across city agencies, keep citizens informed and give them access to services.
<number>
<number>
Winn-Dixie is a Jacksonville, Florida-based grocery chain with more than 600 retail locations throughout the southeastern U.S. Stores range from 25,000 to 50,000 square feet. Faced with an aging IT infrastructure with one server supporting each retail store, Winn-Dixie’s distributed environment was costly to maintain, and individual stores could experience extended outages. For example, when one server went down, a server often had to be sent from a warehouse in another state and then brought up by store staff members. As a result, store outages could last for days.
Shopping for the perfect solutionA long-time IBM Power Systems™ client with key store applications running on AIX, Winn-Dixie investigated both the replacement of individual store servers and the implementation of a consolidated solution in a central location. The substantial cost savings, management simplicity, and redundancy options afforded with a centralized data center approach led the team to consolidate more than 600 servers into three IBM BladeCenter H chassis with JS22 blade servers, each supporting 16 individual stores using PowerVM. “We decided on the IBM BladeCenter using the JS22 blades because of the capabilities that come with that package—including virtualization and redundancy—because uptime is very important,” says Barry Kirk, director of architecture for Winn-Dixie. “We thought the IBM solution gave us the most flexibility not only for now, but going forward. It accommodated our existing application portfolio while providing a very rich upgrade path in our infrastructure.”Savings all across Winn-DixieBy consolidating onto JS22 blade servers, Winn-Dixie avoided more than US$5 million in capital costs. They eliminated the need for staff members to perform server administration at each store, as well the associated energy costs for each location.“All of that saved capital went back into our store remodeling program. So it made what would’ve been an expense into a revenue-generation proposal,” says Charlie Will, CIO for Winn-Dixie. “And our daily maintenance has been reduced considerably because we’ve cut out all of the maintenance that went into those stores. The labor savings is huge, and power usage is greatly reduced.”Simplifying for better manageabilityCentralizing all store processing into a single data center has provided Winn-Dixie with a more dynamic infrastructure. IT staff members at the centralized data center in Jacksonville can now more easily manage the processing needs for all stores through the built-in PowerVM virtualization capabilities on the JS22 blade servers. “In the old days, two years ago, when we wanted to add a new store or remodel a store, we had to procure equipment,” explains Gary Klingerman, senior manager of platform engineering at Winn-Dixie. “We had to send people to the store to install the equipment, operating systems and applications. Now when we add a store, we simply create a new LPAR on our JS22 blades, and we’re ready and cooking.”Implementing a comprehensive disaster recovery strategyWinn-Dixie is currently running a total of 39 BladeCenter JS22 blade servers with 16 store locations on each blade in the BladeCenter H chassis. There are three blades (one per chassis) that serve as hot standby blades to be used in case of failures.With its centralized data center in Florida, the company faces the potential threat of hurricanes wiping out their entire IT infrastructure. Winn-Dixie’s first tier of disaster recovery is an exact mirror of the headquarters solution at a nearby facility in Baldwin, Florida. Integrated tagging ensures that all LPARS are running with the exact same IP used at headquarters. As soon as IT staff members make the network changeover, they can be up and running at the Baldwin facility.Winn-Dixie also has a second tier of disaster recovery that consists of this BladeCenter solution at a facility in Gaithersburg, Florida. This location is also on standby in case the company should lose all computing capabilities at both the Jacksonville and Baldwin facilities.Gaining a competitive edge with high performanceSince the implementation of the JS22 blade servers with IBM POWER6™ processors, Winn-Dixie has benefited from a significant performance increase over their previous systems. “We’ve made our data centralized, and our performance has gone up five to ten times on our applications in the store,” says Klingerman. “We have some applications that used to run in excess of two hours. They now run in around ten minutes on the JS22 blades. Store associates that are using handheld inventory scanners are getting faster response times, so they have more time on the sales floor to provide service to customers.”A reliable solution from a trusted partnerWinn-Dixie IT staff members relied upon the partnership they have with IBM to provide the best answer to their business challenge. “IBM understood the business goals we were fighting for. We were not just getting a piece of hardware,” says Will. “They were instrumental in helping us come up with the components to make this solution work holistically.”
<number>
Baylor College of Medicine (Baylor) was building a new hospital as an adjunct to its medical school and research and clinical facilities. Its plans included upgrading its existing data center, thereby easing space constraints and at the same time, replacing end-of-life Sun servers. A number of vendors were considered, but IBM was selected for its IBM Power servers and its data migration expertise, which includes the ability to maintain constant access to Baylor's SAP ERP system, medical records and billing systems, patient management systems, as well as systems that support its research facilities, grant management and e-mail systems. "We are responsible for the infrastructure to support all these missions, and we take that very seriously," says Jenifer Jarriel, vice president of IT and chief information officer of Baylor. "When we look at partnerships, we look at companies that have the same kind of alignment with our customer service perspective on excellence."
Back to top
Solution implementation
IBM Premier Business Partner Mark III was engaged and brought in the IBM Migration Factory to provide the transition of workloads from 32 legacy Sun servers onto three IBM Power 570 servers, running the IBM AIX operating system. Al Reineking, executive director of IT operations and technical services at Baylor, said, "When the rubber hit the road, the Migration Factory produced exactly what they said they would with no problems at all. They took the pain and more importantly the risk out of migrating very sophisticated software applications from one hardware platform to a new one. In reality, it was a piece of cake."
Back to top
Benefit of the solution
Immediate benefits ensued. There was a 60 percent reduction in footprint relative to the rack space of the SAP environment and more than 40 percent savings in both cooling and power. The consolidation of 32 legacy Sun servers onto the three IBM Power servers and the virtualization of a variety of core business applications enabled the college to reduce its data center footprint from eight racks to three. Managing and maintaining the three new servers has proven to be dramatically simpler. The new ease of management has enabled Baylor to focus on more high-value projects, including making preparations to extend IT services to accommodate the new hospital.The critical nature of patient records and research meant that the new solution needed to be able to provide services to the college's client base around the clock. The IBM solution helps provide fault tolerance and high levels of redundancy to minimize single points of failure in the infrastructure. It also enables capacity on demand to help meet changing processing requirements. "I feel much better about the resources that we're using and the efficiency that we've been able to create," adds Jenifer Jarriel. "And we will continue that direction - we will continue to look at minimizing our footprint and gaining all the benefits we can from virtualization."
Solutions/Offerings
Hardware:
Power Systems: Power 570, Power Systems running AIX
Solutions:
ISV/Alliance Partners: SAP ERP
<number>
This case study shows how The Coca-Cola Bottling Company of Egypt implemented two IBM Power ™ 570 servers running IBM AIX®, IBM PowerVM™ Dynamic Logical Partitioning and Micro-Partitioning, and IBM System Storage™ DS4800 to improve performance, scalability and availability.
The Coca-Cola Bottling Company of Egypt's legacy hardware infrastructure, based on Competitive servers, hosted the company's Enterprise Resource Planning solution (Oracle e-Business Suite). However, due to an increase in the business expansion and implementation of new modules, the servers failed to perform satisfactorily and adequately meet business needs.The Coca-Cola Bottling Company of Egypt wanted to replace its legacy infrastructure with a high performance, scalable hardware solution. The company also wanted a centralized data storage system to reduce its administration work, while improving its performance.
Back to top
Solution implementation
The Coca-Cola Bottling Company of Egypt chose IBM Power Systems and IBM System Storage technologies to upgrade its server environment. The company implemented two IBM Power 570 servers running AIX, one for the main site to host its production environment, and the other at the disaster recovery (DR) site. The Coca Cola's core business applications are now running on these servers. The company leveraged IBM PowerVM Dynamic Logical Partitioning and Micro Partitioning and implemented it on both servers. It used different logical partitions (LPAR) at the main site server to leverage the flexibility of partitioning, while minimizing server administration. Coca-Cola also implemented PowerVM Micro-Partitioning at its disaster recovery site to decrease the number of used processors to the minimum acceptable, since this server is only used in case of disaster and Testing.IBM System Storage DS4800 was used at the main site as well as the disaster recovery site, to store all the data which includes the company’s production, financial and sales data as well as the ERP solution (Oracle e-Business Suite). This IBM technology provided a robust and highly-available storage environment, ensuring that the Coca-Cola company's data is secure.The IBM Rational Performance Tester 7.0.1.1 together with our European benchmark center was chosen to assess the infrastructure capabilities to ensure that the solution meets the performance required by Coca Cola. Five predefined scenarios, representing the Coca-Cola's core business, were captured and rerun. 1200 "heavy users" connected to the application were executed to assess the solution performance. Rational Performance Tester with the BSD extensions allowed the benchmarking facility to simplify load testing by capturing and managing workload against Coca-Cola's application. The team was able to record and playback test scripts, drive different types of user data against the application, define various user loads, set verification points to validate application usage, and analyze the various performance reports in terms of throughput and user load. Metrics were collected to evaluate the scalability of the proposed platform.
Back to top
Benefit of the solution
By implementing IBM Power 570 servers running the AIX operating system, IBM System Storage DS4800 and IBM PowerVM technologies, The Coca-Cola Bottling Company of Egypt replaced it legacy infrastructure with a flexible solution that supported its ERP solution, providing outstanding performance and high scalability.Some of the other benefits are:- IBM Power 570 servers decreased the server utilization to 65% supporting 600 Users instead of 100% with 350 users on the replaced HP Systems.- Reduced the total number of processors from 16 competitive processors to only 7 IBM Power6 processors.- IBM Power 570 servers running AIX ensure support for the company’s future growth, which is expected to double by the end of 2009.- Consolidating the servers on to one server with logical partitioning enhanced the system management and administration.
Solutions/Offerings
Hardware:
Power Systems: Power 570, Power Systems running AIXStorage - Disk: Storage: DS4800
Software:
PowerVM: PowerVM Dynamic Logical Partitioning, PowerVM Micro-Partitioning
Solutions:
ISV/Alliance Partners: Oracle EBS Other
<number>
University of Pittsburgh Medical Center Rewriting the rules on IT investment to facilitate tomorrow’s healthcare innovations
Business Challenge:
University of Pittsburgh Medical Center’s (UPMC) predominant focus is on transforming its entire IT infrastructure to lay the groundwork for the future, an effort that is far reaching in scope and subject to major challenges—most or all of which are shared by major healthcare providers today. UPMC, Pennsylvania’s largest integrated healthcare delivery network, sought to lower the cost and complexity of IT infrastructure to enable the continued investment in next-generation clinical systems and to lay the foundation for the best possible patient care.
Solution:
UPMC joined with IBM in an 8-year, $402 million partnership designed to transform its IT infrastructure. The deal was viewed as a watershed in how IT vendors and their customers work together. The most telling indicator of the project’s success is the difference between UPMC’s actual capital and operating costs (related to IT) and those that would have been incurred had it taken no action. A key backdrop for this comparison is the surge in processing and storage volume that resulted from the faster-than-projected expansion in the scope of UPMC’s industry-leading electronic health records adoption. Had no action been taken—UPMC would have needed to more than double its number of servers. Instead, it was able to reduce the number of servers by two thirds, and the consequent reduction and avoidance of $30 million in capital and operating costs.
By simplifying its IT infrastructure through virtualization, UPMC is able to support more server capacity without the need to hire any additional support staff. On top of that, the server consolidation afforded by its virtualization strategy enabled UPMC to reduce its floor space requirements.
The transformation making these benefits possible is being implemented by IBM Global Technology Services. The modular, standardized nature of the solution means UPMC can integrate new acquisitions into its network faster—enabling the more prompt realization of the acquisition’s operational and clinical goals.
Benefits:
$30 million in capital and operating cost reductions through virtualization-driven efficiencies
150 percent increase in processing capacity with no increase in IT support costs
40 percent reduction in IT infrastructure floor space requirements, freeing up space for revenue generating services
67 percent reduction in number of physical servers
Expected increase in average utilization per server from three percent to nearly 80 percent
Faster integration of acquired healthcare operations
Why it Matters::
By transforming its IT infrastructure through consolidation and virtualization, UPMC has achieved more than a quantum improvement in resource efficiency. It has fundamentally changed the link between processing and resource needs—enabling it to meet an ambitious clinical agenda with a far lower rate of IT investment growth.
Usage level: Win/Ongoing project/Completed project:
External Complete
Industry:
Healthcare
<number>
<number>
Let me summarize why AIX is the premier UNIX platform today.
Our concurrent development of AIX with the POWER microprocessor, system hardware and firmware drives a level of innovation and performance that the competition simply can’t match. Solaris and HP-UX development and support are split across multiple platforms whereas we are highly focused with a premium offering.
AIX and POWER deliver sheer performance in production environments that the competition cannot match. Our leadership across dozens of industry benchmarks validates this point.
AIX with Advanced Power Virtualization delivers tremendous flexibility and drives workload consolidation for higher system utilization. Virtualization is clearly a strong trend in IT and AIX is very well positioned.
Finally, a clear differentiator is our stable AIX roadmap. I’ll go through our roadmap in more detail in a few slides, but let me say that AIX has a predictable roadmap well into the next decade. We proactively communicate the timing of our releases and have established guidelines for lifecycle planning. This makes life much easier for customers, especially those who are wrestling with the uncertainty of Sun and HP roadmaps.
Now, let’s talk about where AIX is going. Please turn to the next slide.
Let me try and summarize why AIX is the premier UNIX platform today.
Our concurrent development of AIX with the POWER microprocessor, system hardware and firmware drives a level of innovation and performance that the competition simply can’t match. Solaris and HP-UX development and support are split across multiple platforms whereas we are highly focused with a premium offering.
AIX and POWER deliver sheer performance in production environments that the competition cannot match. Our leadership across dozens of industry benchmarks validates this point.
AIX with Advanced Power Virtualization delivers tremendous flexibility and drives workload consolidation for higher system utilization. Virtualization is clearly a strong trend in IT and AIX is very well positioned.
Finally, a clear differentiator is our stable AIX roadmap. I’ll go through our roadmap in more detail in a few slides, but let me say that AIX has a predictable roadmap well into the next decade. We proactively communicate the timing of our releases and have established guidelines for lifecycle planning. This makes life much easier for customers, especially those who are wrestling with the uncertainty of Sun and HP roadmaps.
Now, let’s talk about where AIX is going. Please turn to the next slide.
<number>
Now let’s look at reliability, availability, and serviceability. A recent survey of 400 IT users worldwide by ITIC showed that the combination of AIX and Power Systems provides the best result in each of these categories.
Our availability is 99.997% - 2 ½ times the next best UNIX alternative and 10 times better than Windows on x86.
AIX/Power reliability was best with less than one outage per year per system. This is particularly impressive when you consider that Power systems dominates the midrange and highend and so the average configuration of our systems is much larger than the competitive systems. We have more hardware to fail but it doesn’t bring the system down because of the way our HW, firmware, and OS work together.
This survey only measured one aspect of serviceability – patch time to the OS – and AIX on Power was the best at that as well with only 11 minutes average time to install a patch.
54% of IT execs surveyed say they need 99.99% availability or better. With these kind of results it is no wonder that more and more of them are choosing Power systems.
Let’s go to next chart and lok at why this is happening.
<number>
Over the years AIX has evolved from an engineering workstation operating system to a solid dependable server operating system
<number>
POWER Hardware has an optional, separately chargeable hardware feature called “PowerVM” that enables some of the advanced virtualization capabilities of the POWER hardware including:
- Enables (turns on) Micro-Partitioning
- Includes new support software, the Virtual I/O Server
- Includes automated load balancing software, the Partition Load Manager
PowerVM is managed through IBM Systems Director
<number>
AIX 7 builds on the capabilities of previous releases of AIX, like AIX 6, to provide new capability to clients we are also announcing a new way for clients to consolidate old workloads on AIX 7, with the new product AIX 5.2 WPARs for AIX 7
<number>
AIX can’t continue to grow without providing even more value to our clients. That is why IBM is providing a preview of the next version: AIX 7.
AIX 7 represents over three years of work by the AIX development team. Although there are enhancements across the entire operating system – four key areas are highlighted in this release:
Vertical scalability, Virtualization, Resiliency and management automation. Many of the features of AIX 7 are also available in AIX 6 technology level 6
<number>
IBM issues a guarantee that if your applications are will run on AIX 7 if they ran on previous releases of AIX assuming reasonable coding practices are used
As stated earlier, AIX 7 will be binary compatible with applications created on AIX 5L and will run on Power4 and later systems.
To smooth the transition to AIX 7 we have a number of activities planned
- First, we are previewing this release with our clients months before it becomes generally available
- Second, we have an Open Beta that will allow virtually any client to download and test AIX 7
- Third, we have extensive activities underway with our ISVs to insure that they understand the AIX 6 is binary compatible and to help them exploit some of the new features of AIX 7
- We also have a number of other activities, both for clients and ISVs to introduce this new version of AIX
We believe that these activities will increase the adoption of AIX 7 by our clients and ISVs
As you can see from the slide, clients with a software maintenance agreement will be able to upgrade to AIX 6 at no added charge when it becomes generally available.
Finally, the upgrade process for AIX7 will be exactly the same as if it were named AIX 6.2 – existing tools such as alt disk install can be used to perform a straightforward upgrade to AIX7 with a minimum of disruption and risk.
<number>
Over the past 20 years, AIX has moved from processing a single thread of work at a time, to being able to handle 1024 threads of work simultaneously running – delivering the capabity to handle the largest workload required by your business.
<number>
6C5 – AIX TB Segment Support (AIX 7.1 and 6.1L)AIX V7.1 and AIX V6.1 with the 6100-06 Technology Level introduce a new memory scaling feature that exploits the Power MMU and reduces SLB misses. Large memory workloads running on Power 7 systems will benefit the most from this feature. Not enabled by default on AIX 6 Tl6.
0DE – AIX kernel memory pinning (AIX 7.1 and 6.1L)AIX V7.1 pins kernel memory by default and includes support allowing applications to pin their kernel stack. Pinning kernel memory and the kernel stack for applications with real-time requirements can provide performance improvements by ensuring that the kernel memory and kernel stack for an application is not paged out. This feature can also be enabled in AIX version 6.1 with the 6100-06 Technology Level.
117 - Enable hardware acceleration support for Encrypted File Systems, IPSec and Trusted Execution (Saurabh Desai) (AIX 7.1 and 6.1L)
The state of the art security technologies implemented in the earlier versions of AIX namely the Encrypted File Systems (EFS), IP security (IPSec) and Trusted Execution (TE) use software-based cryptographic algorithms. The AIX Cryptographic Framework (ACF) is a cyrptographic provider in AIX. This feature enables these key security technologies to exploit the hardware cryptographic acceleration facility provided by the ACF.
7K4 – LVM Flash Support (AIX 7.1 and 6.1L)AIX V7.1 includes enhanced support in the AIX Logical Volume Manager (LVM) and JFS2 filesystem for Solid State Disks (SSDs). This includes the capability for LVM to restrict a volume group (VG) to only contain SSDs and the ability to report that a VG only contains SSDs. JFS2 has been enhanced with the capability to capture and report per-file statistics related to hot-file detection that can be used to determine if a file should be placed on an SSD. These capabilities allow for applications to monitor and determine optimal file placement. This feature is also available on AIX V6.1 with the 6100-06 Technology Level. The changes made to LVM are just that; changes to various commands that allow a system administrator restrict a VG to only contain PVs whose underlying media (disk) is SSD. We put this in specifically to support requests from DB2 to be able to determine if a VG is of type SSD (which is supported via lsvg and programmatically). Thus there are changes to mkvg, chvg, and lsvg to support this functionality.............doing a search on the 7.1 pubs for "SSD“To generate the hotness report, a new option “–O hot” needs to be
added as an argument to filemon command.
– Note that this is supported in Automated offline and manual offline modes only.
– For example: To generate hotness report in automated offline mode,
run the following command:
Filemon –O hot –A –r fmon –o fmon.out “sleep 20”
092 – Shared Memory Interface to Barrier Synchronous Register SR (AIX 7.1 and 6.1L) AIX V7.1 provides support for accessing the POWER7 Barrier Synchronous Register (BSR) via shared memory interfaces from user space. This allows utilization of the BSR without requiring a kernel extension to be loaded. This support is also available on AIX V6.1 with the 6100-06 Technology Level.
<number>
What is it
Virtualized AIX operating system environments within a single AIX image – each with unique network addresses.
Each WPAR shares the single AIX operating system but can be separately managed
Applications and users inside a WPAR cannot affect resources outside the WPAR
Each WPAR can have a regulated share of processor, memory and other resources
Two types of WPAR
System WPARs have separate security and appear like a completely separate OS
Application WPARs are manageability wrappers around a single application
# WPAR takes seconds to create and LPARs minutes
# LPARs need setting VIOS LV or LUNs but WPARs add simpler NFS mount points, or can just use Global AIX diskspace options
# LPAR needs 512 to 1GB to boot AIX and a WPAR takes just ~60 MB (yes sixty megabytes)
# You can share application code say 1 GB in each and every LPAR (40 LPAR = 40 GB) or just one shared read-only copy for all WPARs 40 WPAR = 1 GB).This saves man-power in maintenance, disk space AND memory (if the application is loaded in the Global AIX then there is only one copy in RAM)
# Maintenance of one Global AIX plus syncwpar command is much less work than say updating 40 copies of AIX
# Especially if you run WPAR buckets:This means instead of running say 40 LPARs you run three Global AIX LPARs for different 3 AIX 6 Technology Levels that you need - WPARs can be move between these as they are updated to high TL's. Eventually, you empty the lowest one and remove it. Then create a new Global AIX for the next new TL as it becomes available
# Application mobility much simpler to organise than LPM
# Global AIX admin can see and change ALL the filesystems of the WPARs - for example, adding a tool to /usr/local/bin is a simple "cp" command away.
# Lots of WPARs on one AIX is simpler to monitor and control than monitoring across many LPARs.
# Rapid cloning is easy and lets you use "disposable images" - simple to create, experiment and throw away
# If you mess up a WPAR, you can enter it from the Global AIX to fix it - if you mess up an LPAR it may not boot!
# Backups are much easier and smaller than a LPAR mksysb - a default WPAR is around 75 MB as a backup file. Of course, more if you have applications + data but the 2 GB of AIX does not have to be included in the backup
<number>
This slide shows the use of Workload Partitions within Dedicated LPARs and MicroPartitions. The leverages the best
capabilities of each technology. We expect that this will be the dominant usage model for WPARs since it leverages the advantages of hardware isolated LPARs and virtual I/O with the flexibility and potential administrative savings of WPARs
For example, the Finance LPAR needs the highest degree of isolation and the workload demands
a dedicated set of processor resources.
The Planning workload also requires a dedicated number of processors, but the adminstrator has decided to create a System WPAR for the Business Development team – maybe they require root level authority to install new software. By setting up a WPAR for the Business Development workload, the administrator could have reduced her workload because she doesn’t have to be involved in the day to day activity of managing the WPAR. Because all the “Finance workloads” are using a single AIX image, there is less system administration work required to keep AIX up to date.
Another example is in the “Americas” LPAR. The administrator has created two WPARs one for the Manufacturing operation and one for the Planning team. The administrator might have created these two WPARs just so that he can create billing records to capture the amount of resources each one of those workloads consume, for chargeback or capacity planning purposes.
As you can see, clients can use the combination of LPARs and WPARs to improve efficiency. You don’t have to use LPARs and WPARs together, but it does make sense in many cases.
<number>
Workload Partitions not only enables administrative efficiency by potentially reducing the number of AIX images to maintain, but that same technique can also be applied to applications that support a shared installation. By installing the application into the global instance and then sharing the application files (in read only mode) with WPARs, potentially multiple independent instances of the application can be run in separate WPARs.
<number>
6ET – Support for export of Fibre Channel adapters to WPARs (AIX 7.1 Only)
AIX V7.1 includes support for exporting a virtual or physical fibre channel adapter to a WPAR. The adapter is exported to the WPAR in the same manner as storage devices. When the WPAR is started, all child devices will be discovered and configured inside the WPAR, which provides several advantages. SAN devices can be directly provisioned to the WPAR's N-port with no need to provision to the Global partition first and then export to the WPAR. Multipath management can be done within the WPAR. Fibre channel tape systems using the "atape" driver are also supported inside the WPAR in this configuration.
Whether it is a physical FC adapter or a virtual adapter such as a NPIV FC exported from a VIOS, this is what is exported to the WPAR, and the associated WWPN goes with it. Any LUN zoned to that WWPN will be visible to the WPAR and not the Global.
0ZY – VIOS disk support in a WPAR (AIX 7.1 & 6.1L)
With this release of AIX, VIO SCSI disks are now supported in a WPAR in the same manner as Fibre Channel disks. VIO SCSI disks configured in the Global partition can be exported to a WPAR for use as a data disk or as the rootvg for the WPAR. Any disk type supported by the VIOS can be exported in this manner to a WPAR. In addition, Live Application Mobility is supported for a rootvg WPAR on VIOS SCSI disks as long as the source and target LPARs are both clients of VIO server(s) with access to the disks.
6E6 - Trusted Kernel Extension Loading and Config from WPAR (AIX 7.1 Only)
AIX V7.1 provides the capability for a Global administrator to export specific kernel extensions for a WPAR administrator to have the ability to load and configure from inside the WPAR. A signature for the specified extensions is computed at the time of export to ensure only the exported extensions can be loaded. The exported extensions can be specified as to whether they are loaded in a local WPAR namespace, or globally, and therefore accessible by the global and other WPARs. This capability removes a barrier for some application stacks that require a kernel extension so that they can function in a WPAR environment. For AIX V7.1, Live Application Mobility for a WPAR with an exported kernel extension is not supported.
<number>
<number>
<number>
<number>
<number>
Live Application Mobility allows you to move workload partitions from one system to another without restarting the application. This enables greater application availability by allowing you to move workloads away from a system that has a planned outage. It can also be used for workload balancing.
<number>
Live Partition Mobility also allows you to relocate an entire presentation including the operating system from one system to another.
Live Par-
tition Mobility provides the ability to relocate a running LPAR from one
physical machine to another physical machine without disrupting the
LPAR's operation. Live Partition Mobility provides clients significant
flexibility in their management of workloads. Live Partition Mobility
can be used to balance loads between physical servers without incurring
downtime. It can also be used to move LPARs off of a server in order to
perform planned maintenance on the server without incurring an outage.
Live Partition Mobility can also improve system availability by allowing
LPARs to be moved off of a server in response to a pending system out-
age.
<number>
<number>
The IBM Journaled Filesystem Extended (JFS2) provides for even
greater data security with the addition of a new capability to en-
crypt the data in a filesystem. Clients can select from a number of
different encryption algorithms. The encrypted data can be backed up
in encrypted format, reducing the risk of data being compromised if
backup media is lost or stolen. The JFS2 encrypting filesystem can
also prevent the compromise of data even to root-level users.
<number>
The AIX Security Expert was introduced with Technology Level 5 up-
date to the AIX V5.3 OS and provides clients with the capability to
manage more than 300 system security settings from a single inter-
face and the ability to export and import those security settings
between systems. AIX 6 includes an enhancement to the Security Ex-
pert to store security templates in a Lightweight Directory Protocol
(LDAP) directory for use across a client's enterprise.
<number>
AIX 7.1 (Standard Edition and above) supports built-in clustering capabilities. Administrators can use the operating system provided primitives to cluster a set of AIX nodes and take advantage of the capabilities of the clustering. Some of the functions supported by Cluster Aware AIX include:
Commands and APIs to create the cluster across a set of AIX systems: Kernel based heartbeats and messages provide a robust cluster infrastructure and by default exploit multi channel communication between the nodes using the network and SAN physical links.
Advanced cluster-wide event management: AIX event management is enhanced to support cluster wide event notifications for certain events (for example, Network/disk errors).
Global Device View support: When managed by Cluster aware AIX, device files associated with the disks shared across the nodes in the cluster will have a common name across the nodes in the cluster that have access to the disks.
Cluster wide command operation: Many of security and storage related AIX commands have been enhanced to support the operation across the cluster
These clustering capabilities are also available on AIX V6.1 with the 6100-06 Technology Level when using PowerHA SystemMirror (formerly known as HACMP).
<number>
Although new releases of PowerHA SystemMirror and PowerVM VIOS will directly use the Cluster Aware AIX interfaces, existing programs that use the legacy RSCT (Reliable Scalable Cluster Technology) interfaces will continue to work unchanged.
<number>
Nonintrusive Service Aids: AIX V6.1 serviceability aids are designed
to minimally impact performance and availability. Second failure
data capture (SFDC) technology involves building diagnostic and data
capture features into the product code, but only enabling them after
problem diagnosis has started. These aids are tunable and can be localized to specific subsystems, further mitigating any performance
impact. The result is faster, less-disruptive problem determination, without the need to install special debug code. AIX 6 also
introduces a mainframe-inspired live dump facility. This allows se-
lected subsystems to dump their diagnostic information for subse-
quent service analysis, without requiring a full system dump and
partition outage.
AIX V6.1 continues
to build upon the first failure data capture and nondisruptive
service aid features introduced in prior AIX releases. A new
live dump feature allows selected subsystems to dump their mem-
ory state and traces to the file system for subsequent service
analysis, without requiring a full system dump and outage. As
for those problems that still require a partition restart in or-
der to recover, AIX V6.1 provides a firmware-assisted dump mode.
In this new mode, AIX cooperates with system firmware to write
the FFDC information to the dump device using the restarted AIX
image, rather than writing to the dump device using the failed
AIX image. The intended result is fewer dump failures and a more
reliable system dump facility.
<number>
POWER6 Storage Keys: POWER6 Storage Keys exploitation of the
POWER6 processor storage key feature brings a mainframe-inspired
reliability and capability to the UNIX market for the first
time. Storage keys can reduce the number of intermittent outages
associated with undetected memory overlays inside the kernel.
Applications can also use the POWER6 Storage Key feature to in-
crease the reliability of large, complex applications running
under the AIX V5.3 or AIX V6.1 operating system.
<number>
Dynamic tracing with Probevue: AIX V6.1 provides a new dynamic
tracing capability that is designed to simplify debugging com-
plex system or application code without requiring code changes
and recompilation. This dynamic tracing facility will be intro-
duced via a new tracing command, Probevue, that allows a devel-
oper or system administrator to dynamically insert trace
break-points in existing code without having to recompile the
code. A developer or system administrator can use Probevue to
dynamically place probes in existing code, and specify the data
to be captured at probe point.
<number>
<number>
AIX V6.1 provide a new graphical systems console,
the IBM Systems Director Console for AIX. The Systems Director Con-
sole for AIX provides a responsive Web access to common systems man-
agement tools such as the Systems Management Interface Tool (SMIT)
and offers integration into the IBM Systems Director. The Systems
Director Console for AIX is included with AIX V6.1.
The IBM Systems Director Con-
sole for AIX is built on a lightweight infrastructure and provides
an easy-to-use interface for the Web-enabled administration of AIX
management tasks. The console provides a single user interface for
system management and administration operations and can be accessed
from any supported Web browser. IBM Systems Director Console for
AIX gives the user a powerful interface to manage AIX servers and
software and provides a graphical Web-based interface to enable the
user to manage remote systems and resources.
The IBM Systems Director Console includes these features:
o AIX V6.1 provides browser-based access to the popular System
Management Interface Tool (SMIT). Users can access AIX operating
system management functions in a Web browser.
o Distributed Command Execution Manager provides the capability to
securely execute systems management commands on multiple sys-
tems. AIX V6.1 includes all the infrastructure needed for the
IBM Systems Director Console for AIX. The Console is not a
prerequisite for IBM Systems Director, but it is designed to
have a similar management interface and shares a common technol-
ogy base.
<number>
With Automatic Variable Page size support, AIX will automatically select
optimal page sizes that will result in better performance for some ap-
plications. All of this is done transparently to the applications and
does not require any application changes or recompilation. AIX will
choose optimal page sizes for an applications data, heap, stack, shared
memory, and anonymous mmap() memory dynamically based on reference and
use patterns. POWER6 systems are required for this feature, as AIX ex-
ploits a new capability of the POWER6 processor to dynamically promote,
demote, and mix 4 KB and 64 KB page sizes within the same memory region.
This feature is active by default and includes system administrator
controls to disable or to tell AIX to be even more aggressive with this
new page promotion feature.
<number>
AIX Profile Manager is an Advanced Manager of IBM Systems Director 6.2. It is provided as a feature of AIX, distributed with Standard and Enterprise Editions. AIX Profile Manager provides "dashboard" views and runtime configuration profile management over groups of networked systems. The profile management capability includes capturing the runtime configurations, applying the new runtime control values, and comparing the run-time configurations with a given profile or between systems. AIX Profile Manager manages the client AIX systems using the AIX Runtime Expert technology. It also provides "copy machine" like simplicity for distributing OS properties to many systems. Profile Manager installs on all supported Director platforms (Windows, Linux, AIX). The Profile Manager leverages the AIX Runtime Expert infrastructure
First introduced with AIX 6 Technology Level 4. AIX Runtime Expert provides a simplified solution for managing runtime properties of one or more AIX instances. AIX Runtime Expert in AIX V7.1 provides several enhancements, including improvement in performance, ability to set the version of the configuration profiles to maintain multiple versions, ability to set customized descriptors to the profiles. It also provides the ability to be managed by AIX Profile Manager from an IBM Systems Director. This feature is also available on AIX V6.1 with the 6100-06 Technology Level.
<number>
<number>
As you can see from this chart, there are a number of cases where a customer can use either LPARS/MicroPartitions or Workload Partitions.
IBM is offering clients additional capability that they can choose to use when appropriate. Feedback from our clients indicates that many intend to
use WPARs and LPARs in their environment. We understand that there isn’t a single solution to every problem but that by providing customers
with a broad range of virtualization capabilities across our entire product line, out clients will have the tools to more efficiently run their IT operations.
<number>
AIX 6 includes many important security capabilities, but remember, if your client wants to manage security on systems other than just AIX, or they want to take an enterprise approach to managing security even just on AIX – the client should be made aware of the numerous security management offerings from Tivoli including Tivoli Access Manager.
<number>
<number>
This chart summarizes some of value provided by select AIX 6 features and any particular segments that might be particularly interested in that feature. As you can see, most of the functionality provided by AIX 6 are of general interest to all of our clients.
<number>
AIX Enterprise Edition is a single offering that includes AIX 6 and several key manageability products that are designed to:
Improve availability
Enhance operational efficiency
Measure resource usage in a virtualized environment
Enterprise Edition consists of:
AIX 6 operating system
PowerVM™ Workload Partitions Manager for AIX (WPAR Manager),
Tivoli® Application Dependency Discovery Manager (TADDM)
IBM Tivoli Monitoring
IBM Usage and Accounting Manager Virtualization Edition for Power Systems.
IBM Service Management is an approach to managing information technology (IT) that views IT resources from the perspective of how those resources support business objectives.
AIX Enterprise Edition provides a sound basis for Service Management by enabling a better understand of how IT components affect the delivery of business services.
The IBM Systems Director family of products are designed to provide consistent platform management across all IBM systems.
AIX Enterprise Edition is designed to complement the platform management capabilities of IBM Systems Director.
The AIX Enterprise Edition is designed to simply the process of managing virtualized AIX environments by including AIX 6 and enterprise management functions into a single, easy to order offering.
By offering AIX 6 in conjunction with enterprise management capabilities, clients can get all of the virtualization, security, availability and manageability features associated with AIX 6, and all the enterprise management features of the WPAR Manager, TADDM, Tivoli Monitoring and Usage and Accounting at an attractive price compared to purchasing the individual products separately.
The AIX Enterprise Edition includes integration between the products such as ITM and TADDM that is not available in the standalone products.
AIX Enterprise Edition is designed to be used in conjunction with IBM Systems Director to provide platform management capabilities.
AIX Enterprise Edition is ideal for clients who are moving up to AIX 6, and adopting PowerVM™ virtualization technologies to control IT complexity and growth.
<number>
AIX Enterprise Edition is designed to improve the manageability of a virtualized AIX environment. Potential benefits include:
Improved service availability through access to relevant, real time information about IT resources
Enhanced operational efficiency though visualization of the relationships of IT resources and applications
Improved operational efficiency by centralizing monitoring and automating the response to service issues
Improved customer satisfaction by providing accurate information of the IT resources used to support business workloads.
Better utilization of IT resources through consolidated views of virtualized resource usage
Improved service availability through Live Application Mobility.
Simplified ordering of AIX and essential service management tools at an attractive price.
<number>
ITM 6.2 gives the Power Systems administrator the ability to be alerted or notified when something goes wrong. ITM uses agent technology, and therefore has the capability to determine the health and availability of the entire Power System, right down to the network interface card.
ITM provides the administrator the ability to monitor both the physical and logical resources of the Power System, including the disk and network that sits behind the Virtual I/O Server (VIOS). It can do this because with every VIOS server shipped from IBM, there’s an ITM agent embedded in the VIOS to enable ITM to monitor the disks and network.
Included with ITM is a data warehouse tool that allows the customer to store as much historical data as they desire. This DB information allows the customer to go back and compare current system performance and utilization to past performance and utilization.
In addition, ITM includes a vast number of reporting templates that can be easily customized for individual customer requirements
<number>
Tivoli Monitoring is virtualization aware – it can show the relationships between virtual resources and physical resources.
Problem:
“Total Monitored Memory Used Pct” value is higher than it should be since it was calculated compared to total “monitored” memory rather than total CEC memory. This will be fixed in an APAR.
<number>
<number>
Out of the box situations:
If the LPAR, nw or storage will reach a limit in 7 days with strength of 3 (confidence + # of samples). Limit for attribute collected and time period
System_pSeries_LogPart_TTCT_1W
System_pSeries_LogPart_TTCT_1W predicts if the monitored logical partition, network or storage will reach a defined critical limit within the next seven days, and the prediction is strong (Strength >= 3) or moderate 2. The limit is set in the analytical task and the period is set in the situation.
System_pSeries_LogPart_TTWT_1W
System_pSeries_LogPart_TTWT_1W predicts if the monitored logical partition will reach a defined warning limit within the next seven days, and the prediction is moderate (Strength >= 2). The limit is set in the analytical task and the period is set in the situation.
System_pSeries_Network_TTCT_1W
System_pSeries_Network_TTCT_1W predicts if the monitored networking will reach a defined critical limit within the next seven days, and the prediction is strong (Strength >= 3). The limit is set in the analytical task and the period is set in the situation.
System_pSeries_Network_TTWT_1W
System_pSeries_Network_TTWT_1W predicts if the monitored networking will reach a defined warning limit within the next seven days, and the prediction is moderate (Strength >= 2). The limit is set in the analytical task and the period is set in the situation.
System_pSeries_Storage_TTCT_1W
System_pSeries_Storage_TTCT_1W predicts if the monitored storage will reach a defined critical limit within the next seven days, and the prediction is strong (Strength >= 3). The limit is set in the analytical task and the period is set in the situation.
System_pSeries_Storage_TTWT_1W
System_pSeries_Storage_TTWT_1W predicts if the monitored storage will reach a defined warning limit within the next seven days, and the prediction is moderate (Strength >= 2). The limit is set in the analytical task and the period is set in the situation.
<number>
TADDM is the product that helps you discover the Power Systems resources in the enterprise.
TADDM is an agent-less technology, that actually uses sensors to go out and discovers all the Power resources and associated applications in the enterprise, and then builds a topography map of the Power configuration. It builds the map, with the associated data from both a physical viewpoint and a logical viewpoint right down to the application level.
It also discovers the relationships between the systems. So if for example you had a mission-critical business application running on a Power 595 that accessed a database running on a Power 560, and you lost the 560, you would be able to discover the full impact of losing that server from a business service perspective.
TADDM also tracks changes down to the file level. A study by IBM determined that 80% of all problems that occurred happened because something in the environment changed. With TADDM you would be able to track all changes, compare those changes with a “Golden Master”, and quickly determine if the source of the problem is the change that was made.
<number>
IBM® Systems Director Standard Edition provides virtualization management for clients that have broadly deployed virtualized resources in a production environment. It allows IT managers to create and manage a standardized set of virtual appliances – virtual machines containing a complete ready-to-run workload comprising OS, middleware and applications – and maintain these within a centralized library or repository. This approach allows a small set of ‘golden master’ virtual appliances to be maintained, which are ready for instant deployment as business needs change. It also streamlines upgrades and patches, thereby eliminating redundancy and ensuring consistency across the virtualized infrastructure.
<number>
In order to respond to changing workloads and business requirements, Power servers have the ability to dynamically optimize your virtual resources. With IBM Systems Director Editions, you can move workloads from heavily loaded servers to servers that have spared capacity. The ability to move running partitions from one server to another offers you the ability to balance these workloads and resources.
For example, if an application’ s resource requirements peak unexpectedly to a point where there is contention for server resources, you might move it to a more powerful server or move other, less critical, partitions to different servers, and use the freed-up resources to absorb the peak. Systems Director Editions provides the simplified management needed to automate this mobility.
<number>
IBM® Systems Director Enterprise Edition adds a new layer of client value by introducing the concept of system pools – groups of virtual appliances deployed across multiple physical servers which can be managed as a single entity. Virtual appliances include servers, storage, and network resources.
You now have the ability to create and remove system pools across multiple physical servers. It also provides a dashboard view of the overall health and status of the pool and deployed workloads along with simplified monitoring and visualization of the aggregate capacity and utilization for the systems within a pool.
Management of system pools also allows relocation of virtual workloads in response to predicted host system failures without disruption. This enables workload resilience for planned and unplanned downtime.
Systems Director Enterprise Edition is designed to help provide increased system utilization, availability, and improved energy efficiencies through pooling and automation.
<number>
Systems Director Editions enables you to monitor and control energy usage on supported systems and Intelligent Power Distribution Units (PDUs).
With Standard and Enterprise Editions:
You can get information on energy usage on single servers or across a rack of servers.
You can get information on thermal characteristics including inlet and exhaust temperatures,
You can also look at trends over time, because a temporary spike in electrical energy use or heat may not be a problem unless it happens frequently or is increasing over time.
You can calculate energy costs for targeted resources
The Energy Cost Calculator allows you to display the amount of energy used for a given resource or group of resources over a specified period of time and calculates the corresponding cost of that energy. It provides a visual indicator of the amount of energy consumed by the resource or group of resources over the specified time period. This is compared to what would have been consumed had nameplate power been drawn over that entire period.
This is just some of the information available that can help you build a more energy efficient IT operations center, in that it helps you to identify potential trouble spots, and to not waste time and resources on areas that are not a problem.
Note that energy management features vary by system.
<number>
Systems Director Standard and Enterprise Editions also provides integration of server, storage, and network management for virtualization environments across platforms. Network management is a key component of an integrated data center management strategy, which is required for a dynamic infrastructure.
These Editions allow you to manage, monitor, and configure vendor based network devices as well as integrate of vendor applications. It also helps protect your IT investment through its open design and support for industry standards, which enables heterogeneous physical and virtual management with support for multiple platforms and operating systems.
The following network devices are supported with Systems Director Standard and Enterprise Editions:
IBM BladeCenter supported switches and Blade NICs
Brocade® Converged Ethernet network switches (FCoCEE) and Brocade® Fiber Channel switches
QLogic® FCoCEE switches and adapters
<number>
<number>
It’s all on the slide.
<number>
Here are some contacts and a sales kit to help you with information on AIX Enterprise Edition .
<number>
<number>
This chart shows the advantages that AIX Express Edition provides when consolidating smaller workloads on larger servers.
Using AIX Standard Edition, a client would have to spend up to $46,800 for AIX licenses for the consolidated environment.
If they use AIX Express Edition the cost of AIX licenses is reduced to $22,800.
US prices are shown, so the difference may be more or less in your country, but the basic message is the same – AIX Express can help you save money!
IBM License Metric Tool is a no-charge offering that is designed to help clients maintain an inventory of Processor Value Unit (PVU) based and per processor core based software licenses. ILMT is designed to assist clients to complying with the AIX Express Edition license requirements restricting the amount of processor capacity in cores available to run AIX Express Edition in an LPAR.
The IBM License Metric Tool has been available for several years and is the strategic license management tool for IBM. Clients that have used IBM software such as DB2 or WebSphere are likely to be familiar with the IBM License Metric Tool.
ILMT is designed to simplify license management by helping clients:
Track and manage processor capacity available to the AIX and Power Systems Software products: Use the reports to help determine if you have the appropriate capacity (Full and Sub-capacity) for each of your Power Systems products including AIX that are installed in your IT environment.
Lower liability risks: Reduce the risk of not meeting your per processor core contractual licensing conditions as well as the unplanned cost of license compliance payments.
Track software inventory: Helps maintain a continuously updated inventory of where IBM software assets are installed in your environment.
For now, in Power Systems Software, the requirement is limited only to clients running AIX Express Edition on a medium (P570, Power 770) or high tier server (P595, Power 780). Clients in this scenario are required to download, install, and use the IBM License Management Tool (ILMT) within 180 days of deployment, to help them manage compliance with the AIX Express Edition license restrictions.
AIX Express Edition is the first product that is specifically limited in vertical scalability by the license. Since there is no hard coded limit inside AIX Express Edition that programmatically prevents it from being used in partitions larger than 4 cores and 8GB of memory per core, it would be easy for a client to unintentionally exceed the license restriction. The IBM License Metric Tool is provided to help clients maintain license compliance. AIX Express Edition is targeted for smaller installations as is reflected in the price point. Thus, it is restricted to availability for 4 processor-core partitions only.
<number>
AIX 6 will support systems based on the PPC970, POWER4, Power5, Power6 and POWER7 processors at all chip frequencies.
Since almost all AIX 6 features will be available on all systems, clients may decide to move up to AIX 6 on older hardware
just to take advantage of the new features such as WPARs, Application Mobility and Role Based Access control.
AIX 6 will only run the 64 bit kernel. This will enable even greater scalability in the future. Just like on AIX 5L, 32 bit and 64 bit
applications will continue to run on AIX 6 – no recompilation is necessary.
Because AIX 6 only includes the 64bit kernel, 32 bit device drivers and kernel extensions will not be supported on AIX 6. Most
device drivers and kernel extensions are already available in 64-bit format, so we are expecting few issues related to the discontinuing
of the 32-bit kernel
<number>
Power7 provides 4-SMT threads; Vector Scalar Extension support; and 32 storage key support
Benefits
Power7 provides 4-SMT threads; Vector Scalar Extension support; 32 storage key support; and P6 Compatibility mode
Benefits
4-Thread SMT thread capability further improves throughput performance and processor core utilization; and AIX/Hypervisor dynamically optimize SMT mode based on load providing better single-thread performance when not all of the SMT threads across all of the cores are needed.
32 Storage Keys Support providing better granularity of storage key isolation providing for 32 concurrent protection regions.
New Vector Scalar Extension (VSX) computation support allowing two independent 2-way SIMD FP operations (double-precision) per cycle, using a 64-entry, 128-bit wide VSR architected register file. Provides enhanced Scalar FP support with a 64-deep register file.
Power 7 AIX Process/Thread Enhanced Memory Affinity
Memory locality provides performance improvements. AIX will take advantage of local/near/far memory hierarchy and CPU affinity will be provided in a micro-partitioned environment.
Additional affinitization of kernel memory will be provided.
Autonomic application affinity improvements.
AIX provides application affinity framework enhancements
Middleware opportunity for improved performance
Extension of existing APIs for better managing affinity
P6 Compatible Mode
enabes support for running prior Technology Levels on Power7
enables support for Live LPAR Mobility from Power6 systems to Power7 systems and vice-versa
64-Core/256-way Power 7 Performance Scaling
Targeted performance scaling improvements for 128/256-way environments
Known scaling issues
128-way performance issues
Analysis, Identification of 256-way performance issues and implementation of targeted improvements
256-core/1024-CPU Enablement in AIX
AIX scales and runs on a fully configured Titan HE (256-cores/1024 SMT threads)
256-core/1024-CPU Power 7 Scaling
Sap 2-Tier Benchmark
Represents a single OS image measure of multiple SAP Application Servers and a single Database server.
Customers use SAP 2-Tier to measure server capacity and use it to construct 3-tier configurations
Customers use SAP 2-Tier as a measure of generic server capacity (beyond ERP)
Database Scaling
Most high end customers deploy SAP in 3-tier configuration
Database scaling work is included to support customer SAP deployment
Benefits
AIX will continue to scale and provide industry leading performance for large workloads.
AIX 6 TL2 and AIX 5.3 TL9 will be supported on any Power7 system that ships in the first half of 2010. These TLs will be supported in P6 mode.They will not be supported on systems that ship the second half because those TL levels will be getting too old.AIX 6 TL4 and up will support P7 mode....prior AIX 6 TLs will not. AIX 5.3 will not support P7 mode, only P6 modeAIX 5.3 will be supported by the Power7 models (at least though 2010) and will be w/d 4/2011AIX 7 will only support P7 mode and extreme scalability (up to 256 cores)Power 7 Exploitation features– Energy Management– Per-Partition energy management– 4 thread Simultaneous Multithreading (SMT)– 1024 Threads / 256 cores partition size– Double precision Vector Scalar Extension (VSX) – 32 Storage Protection KeysAIX version support for POWER7– AIX 6.1 TL2 and later (1st half 2010 only, then TL3 and later moving forward in 2nd half)– AIX 5.3 TL9 and later (1st half 2010 only, then TL10 and later moving forward in 2nd half) – AIX 7POWER7 provides two modes– Power7 mode* AIX 6.1 TL4 and later* AIX 7 – Power6 mode * AIX 5.3 and AIX 6.1 TL3