Aug17 2010 annc faq

IBM Power 795 server, Power 710, 720, 730 and 740 Express servers, and AIX 7

This document is for IBM and IBM Business Partner education only. It is not for client distribution.

Announcement date: August 17, 2010

IBM Power Systems Qs and As

Announcement Overview

These Qs and As provide additional information about many of the IBM announcements made on

August 17, 2010, and related topics.

IBM Power Systems™ server announcements

• Power® 795 – this up to 256-core server provides the performance, scalability, I/O

expandability, RAS (reliability, availability and serviceability) features and energy-

efficient processing required for the most demanding computing requirements.

• Power 710 and 730 Express – these compact 2U (2 EIA unit) 19-inch rack-mount servers

offer a wide range of POWER7® processor options and a rich set of integrated

capabilities. The Power 710 Express provides four, six or eight processor cores in one

socket with up to 64 GB of memory, while the Power 730 Express is a two-socket server

with up to 16 processor cores and maximum memory capacity of 128 GB.

• Power 720 Express – this single-socket server offering four, six or eight POWER7 processor cores, up to 128 GB of memory and extensive I/O expandability is available

either as a 4U rack-mount drawer or in a tower design.

• Power 740 Express – also available as a 4U rack-mount drawer or tower, the Power 740

Express offers one- or two-socket options ranging from four to 16 processor cores, up to

256 GB of memory, many integrated capabilities and significant I/O expandability.

• Power 750 Express processor options – the popular Power 750 Express server, initially

available only in a 32-core 3.55 GHz POWER7 processor configuration, now offers

8, 16, 24 or 32 POWER7 processor cores running at 3.55 GHz.

AIX announcements

• AIX® 7.1 – available in Express, Standard, and Enterprise Editions, the latest release of

IBM’s open standards-based UNIX® operating system offers significant new capabilities

for virtualization, security, reliability and system management.

• AIX 5.2 Workload Partitions for AIX 7 – this new offering allows legacy AIX 5.2

applications to run in a PowerVM™ workload partition running on AIX 7.1.

• PowerHA™ SystemMirror for AIX V7.1 Standard Edition – working with Cluster Aware AIX 7.1, this latest generation of IBM clustering solutions for high availability enables

simpler, more robust cluster formation and management

IBM i announcements

• IBM i Solution Editions – these preloaded solutions provide operating system, database,

security, systems management and implementation assistance and are available for

qualifying ISV applications running on Power 720 Express or Power 740 Express servers.


Annc FAQ.doc Page 2 This document is for IBM and IBM Business Partner education only. It is not for client distribution.

• IBM i support for new Power Systems servers – the Power Systems 795 and Power

Systems 710, 720, 730 and 740 Express servers announced today are supported by

IBM i 6.1 and IBM i 7.1.

• IBM i Edition Express for BladeCenter® S – now available are configurations using the

POWER7 processor-based BladeCenter PS700 Express blade and a choice of SAS storage

switch modules with or without RAID capability.

Integrated solution announcements

• IBM SmartAnalytics System 7700 – this integrated solution, based on several pretested

Power Systems 740 Express server configurations, the AIX operating system and

IBM DB2® database and InfoSphere™ Warehouse V9.7 software, delivers a complete

data warehouse infrastructure for analytics with flexibility for future growth.

• IBM Service Delivery Manager – an IBM statement of direction to extend support to

IBM Power Systems for this integrated software solution designed to help simplify and

accelerate the deployment of a cloud management infrastructure.

Software tools announcements

• IBM Rational® Power Appliance – a new application development environment solution

for Power Systems which includes an IBM Power Express server pre-installed with a

comprehensive set of Rational development software.

• IBM Rational development tools – the Rational C/C++ compiler now supports the Linux®

operating system. IBM plans to add Linux on Power development tools to the Rational

Developer for Power family in the future.

• IBM Tivoli® Usage and Accounting Manager 7.1.3 – the latest release of this Web-based

usage tracking system provides improved support for PowerVM virtualized environments.

Power Systems and BladeCenter features announcements

• PCIe-based SSD (Solid State Drive) offering – this feature consists of a PCIe RAID &

SSD SAS Adapter into which up to four 177 GB SSD modules can be plugged. It offers a

significant price/performance improvement for many client SSD configurations.

• Enhanced I/O offerings for the IBM i operating system – included in today’s IBM i

announcements are:

o Support of the PCIe Cryptographic Coprocessor 4765-001

o IOP-less support of IBM ProtecTIER deduplication and replication technology

o Support of IBM TS2250/TS2350 LTO5 tape drives

o Support through Virtual I/O Server for Fibre Channel over Ethernet (FCoE)

• Improved availability for 128 GB memory feature – availability for the 128 GB memory

feature 5602, which enables the 2 TB memory maximum on Power 770 and Power 780

servers, has been improved to September 17, 2010.

• BladeCenter features – a 2- or 4-port Ethernet expansion card for Power Systems blade

servers and a bridge module to enable convergence into a single network on a

BladeCenter H chassis of the local area network (LAN) and storage area network (SAN).

• IBM 7042-CR6 Rack-mounted HMC (Hardware Management Console) – provides a

dedicated workstation designed for configuration and management of hardware partitions

and Capacity on Demand (CoD) for POWER5™, or later, processor-based servers.

• IBM 7216-1U2 Multi-media Storage Enclosure – this attractively priced 1U storage

enclosure can house two media drives. Drive options are SAS LTO5 tape, SAS DAT320

tape, USB DAT320 tape, SATA DVD-RAM and USB Removable Disk (RDX).



• Low-profile PCIe adapter features – eleven new adapter features for use in the low-profile

PCIe slots on the Power 710, 720, 730 and 740 Express servers announced today.

Index

Power 795 server Page 4

Power 710, 720 730 and 740 Express servers Page 9

Power 770 and 780 server topics Page 18

AIX 7.1 Page 19

AIX 5.2 Workload Partitions for AIX 7 Page 21

IBM Service Delivery Manager & IBM CloudBurst Page 24

IBM Statements of Direction Page 28

Looking for marketing and sales presentations and collateral? Be sure to check out the

Power Systems Master Sales Kit on SSI and PartnerInfo



IBM Power 795 Server Q1: What performance benchmark results are available for the new Power 795 server?

A1: For additional detail about the Power 795 server, including performance data, see:

http://www.ibm.com/systems/power/hardware/795/perfdata.html

For a summary of the characteristics of the Power 795 server, see the “At-a-glance” table at

the end of this section of the Qs & As document.

Q2: What is the significance of the Power 795 server announcement? How does it compare

to the Power 595 server and to competitive systems? Does it replace the POWER6

processor-based Power 595 server?

A2: The Power 795 is the most powerful UNIX system in the industry, offering over 4X the scale

of the Sun M9000 and HP Superdome systems, with dramatically better energy efficiency than

either of them. The Power 795 offers massive throughput, performance and scalability with up to

256 cores of the fastest POWER7 processors available. It is designed for large-scale

consolidation of older Power servers, as well as energy-wasting, under-utilized competitive

servers. The Power 795 enables unprecedented levels of utilization and resource sharing to

support AIX, IBM i and/or Linux applications. It is engineered to deliver the highest levels of

Power Architecture™ RAS (reliability, availability and serviceability). And, available model

upgrades from Power 595 allow clients to leverage their investment in POWER6 systems to

deploy POWER7 performance, scalability and efficiency within their enterprise. The Power 795

is certainly designed to replace the Power 595 offering as the flagship of the POWER7 Power

Systems portfolio. It delivers up to 40% more performance per-core, in TurboCore mode, and up

to 5X more total capacity (rPerf in MaxCore mode) than the 5 GHz POWER6 processor-based

Power 595 server.

Q3: Can a client upgrade his Power 595 to the Power 795?

A3: Yes, the Power 595 (POWER6 based 9119-FHA) may be upgraded to the Power 795 via an

“in-frame” upgrade process that replaces the critical components needed to deliver the POWER7

processor-based Power 795. Clients that have installed a Power 595 have already invested in the

frame, I/O and energy infrastructure of the Power 795. New processor books and DDR3 memory

are delivered in the book replacement that is part of an upgrade to POWER7. Several other

components like the light strip, bulk power controllers, and system controllers are also replaced.

Within a single day, clients may upgrade to POWER7 and enable over 5X the capacity and up to

40% more performance per-core over a Power 595 within the same footprint – leveraging their

initial investment in POWER6, PowerVM, AIX, Linux, IBM i and other software.

Q4: Is it possible for me to upgrade my POWER5 or POWER5+™ p5-595 to a Power 795?

A4: There is no path for a direct upgrade from p5-595 to the POWER7 processor-based

Power 795. Upgrades have been available since the introduction of the Power 595 to enable

p5-595 clients to move forward to the POWER6 processor-based Power 595 (9119-FHA), which

is subsequently upgradeable to the Power 795. Accounting guidelines require a certain

percentage of the system to be preserved as part of a model upgrade in order for an upgrade path

to be offered. From a p5-595 to the Power 795 every major aspect of the system has changed –

the memory, the frame, power subsystem, processors and I/O technology – and the required

percentage of retained components is not met. Clients may replace their p5-595 with a Power 795

or even Power 780 or 770, as appropriate, and deliver significant scalability, performance and

improved RAS while potentially saving on maintenance and significantly on energy consumption.



Q5: Can you upgrade from a Power 780 to a Power 795?

A5: No, there is no upgrade path from a Power 780 to the Power 795. If clients anticipate

needing more performance per-core and/or capacity than provided by the 64-core, 3.86 GHz

Power 780 or the 32-core 4.14 GHz Power 780 system (TurboCore mode), they should consider

initially purchasing a Power 795.

Q6: How is the Power 795 positioned with respect to the Power 780 server?

A6: The Power 795 and Power 780 are both powerful and scalable Power servers, offering

similar value to clients. As enterprise-class offerings, they deliver the same AIX, Linux and

IBM i operating system support, virtualization, TurboCore mode and RAS features – but on a

different scale. The Power 795 offers 4X the scale over the Power 780 in terms of the number of

processor cores (256 to 64 in MaxCore mode, 128 to 32 in TurboCore mode). The Power 795

also offers the fastest processor in the Power Systems portfolio, 4.0 GHz in MaxCore mode

(3.86 GHz on Power 780) and 4.25 GHz in TurboCore mode (4.14 GHz on Power 780).

Power 795 offers 4X more memory (8 TB) than the 780 (2 TB) and 4X the GX++ adapters (32 on

Power 795) for I/O capacity. The Power 795 offers more total bandwidth and capacity overall

and will support higher levels of consolidation with more LPARs – ultimately 1,000 on

Power 795 (IBM Statement of Direction). Finally, one important new RAS feature is being

introduced with the Power 795 – Active Memory Mirroring for Hypervisor.

Q7: What is Active Memory Mirroring for Hypervisor? Is this capability unique to the

Power 795?

A7: Active Memory Mirroring for Hypervisor is a new RAS feature being introduced on the

Power 795 that is designed to eliminate the potential for a complete system outage as a result of

an uncorrectable error in memory. When Active Memory Mirroring for Hypervisor is enabled

(default) the Power 795 system will maintain two identical copies of the system hypervisor in

memory at all times. Both copies are simultaneously updated with any changes. In the event of a

memory failure on the primary copy, the second copy will be automatically invoked and a

notification sent to IBM via the Electronic Service Agent (ESA). This feature is only available on

the Power 795 at this time.

Q8: What are the IBM estimates of relative performance (rPerf) for the new Power 795

A8: rPerf and CPW performance can be found at the Power 795 website under Performance data

at http://www.ibm.com/systems/power/hardware/795/perfdata.html.

Q9: What is the general availability date for the new Power 795 server and MES model

upgrades from Power 595 to Power 795?

A9: General availability is September 17, 2010, for both initial orders and upgrades from

Power 595 (9119-FHA).

Q10: What is the maximum number of LPARs supported on the Power 795? A10: The maximum number of virtual machines supported on the Power 795 currently is 254.

This is planned to increase (IBM Statement of Direction) to 1,000 at a future date.

Q11: What is the IBM SWG PVU (processor value unit) rating for the new Power 795

server?

A11: The PVU rating for the Power 795 is 120, the same as the Power 595 and 780 servers.



Q12: What are the minimum processor book and activation configurations for a

Power 795?

A12: A 3.7 GHz Power 795 system may consist of 1 to 8 processor books with at least 25% of

the processor cores activated (6 per 24-core book). A 4.0 GHz system may be ordered with 1 to 8

processor books and also requires at least 25% of its processors to be activated (8 per 32-core

book). If ordering a Power 795 system to be run in TurboCore mode, then three or more 4.0 GHz

processor books are required (minimum of 96 physical cores). Since at least 25% must be

activated (24 cores out of three processor books) and since TurboCore mode uses only half the

processor cores (48 cores in the case of three processor books), this means that effectively at least

50% must be activated in TurboCore mode (24 or more cores). This 50% minimum activation

requirement in TurboCore mode is necessary to ensure that 25% of the total cores on the system

would be active in the event that the system were IPL’ed at a later date into MaxCore mode.

Q13: Why must at least three FC 4700 processor books be installed on a Power 795 running

in TurboCore mode (4.25 GHz)?

A13: The Power 795 delivers massive scalability within the architecture of the Power 795. Part

of this system’s design is a SMP fabric linking the individual processor books for inter-node

communication. These inter-node links increase as processor books are added to a system. A

two-book system has a single SMP fabric link between the books. A three-book system has three

SMP links, a 4-book system has a total of six SMP links between the books, etc. POWER7

increases the number of processors by a factor of 4X, now offering 32 cores per book versus eight

cores per book for POWER6 technology, and greatly increasing the amount of traffic that can

flow between books over the SMP fabric throughout the system, as processors access memory

and I/O. To ensure balanced and optimal performance in this densely packaged environment, a

minimum number of links are required to support LPAR sizes larger than a single book. These

minimums include three books for 4.25 GHz TurboCore systems and as a result, they may

support a single LPAR of any size, up to the TurboCore system capacity of 128 cores. In contrast,

the 3.7 GHz and 4.0 GHz systems may be configured with 1 to 8 processor books, but require

four books or more to support single LPARs larger than book size (24-core on the 3.7 GHz

system and 32-core on the 4.0 GHz system, respectively).

Q14: What is Power Flex ?

A14: Power Flex is a multi-system Power 795 infrastructure offering to provide a highly available

and flexible IT environment to support large-scale server consolidation and an enterprise’s most

demanding business resiliency objectives. Power Flex is designed to enable clients to use

purchased processor and memory activations more flexibly across a pool of Power 795 systems,

to increase the utility of these resources, and to enhance clients’ application availability.

Power 795 servers in a Power Flex environment are allowed to share large portions of their

virtual processor and memory resources to provide capacity where it is most needed, and to best

support application availability during occasional planned system maintenance activity.

A Power Flex capacity pool of systems consists of two to four Power 795 systems, each with four

or more 4.0 GHz or 4.25 GHz processor books and 50% or more permanent (CUoD) processor

and memory activations to support its applications. Capacity above 25% on these systems may

be used as Flex Capacity Upgrade on Demand resource and re-balanced to/from another

Power 795 system in the same capacity pool up to twelve times per year.



Each Power Flex system also includes a quantity of included On/Off Capacity on Demand

processor days (approximately 60 days of the inactive resources on each purchased Power Flex

64/128-core system, or 960 days per 32-core processor book). These On/Off processor days are

enabled via the normal Capacity on Demand resource enablement process, and may be used at a

client's discretion to provide utility computing for short term projects, workload spikes or in the

event an immediate maintenance activity where an Advanced Planning Event or Re-balancing

request has not been requested.

Power Flex is enabled by ordering RPQ 8A1830 on each of two or more Power 795 systems that

meet minimum requirements and completing the corresponding contract attachments and

supplements for On/Off Capacity on Demand and Power Flex.

Q15: Why are multiple rPerf values given for the Power 795?

A15: Starting with the Power 795 announcement, we are providing multiple rPerf values based

on the size of the partitions. Since most systems of this size will be partitioned, it makes sense to

provide estimates based on the expected client usage. 64-core partition sizes are provided for

reference to the Power 595 system. rPerf values for 16-, 24-, and 32-core partition sizes are

provided based on processor frequency to provide ideal, but realistic, estimates of performance.

Best practices recommend aligning partition sizes with book sizes whenever possible for peak

performance.

Q16: How is CPW measured on the Power 795?

A16: CPW is limited to a 32-core partition size. Systems with more than 32 cores were

measured with multiple partitions. TurboCore results were measured with 16-core partitions.

Q17: What are the characteristics of the IBM Power 795 server? A17:

IBM Power 795 server at a glance General description

Machine type - Model 9119-FHB

Base system form factor 24-inch system frame for CEC + up to three 12X I/O drawers

Optional 24-inch expansion form factor 24-inch expansion frames for additional 12X I/O drawers

Optional 19-inch expansion form factor 19-inch racks for 19-inch I/O drawers

Processor Technology POWER7

Processor packaging 1 to 8 processor books per system

Number of sockets 4 to 32 (4 sockets per processor book)

Processor core options GHz (# cores/socket) | # cores per system

3.7 GHz (6) | 6 a – 192 (standard mode only)

4.0 GHz (8) | 8 a – 256 (standard mode)

4.25 GHz (4) b | 24

a - 128 (TurboCore mode)

EnergyScale™ Y

L2 cache per core 256 KB

L3 cache per core 4 MB (standard mode)

8 MB (TurboCore mode)

Optional Processor Capacity on Demand (CoD) Trial, On/Off, CUoD, Utility

rPerf range c

3.7 GHz (6): 273.51 – 2,188.08 4.0 GHz (8): 372.27 – 2,978.16

4.25 GHz (4) b : 347.36 – 1,852.56

CPW range 3.7 GHz (6): 39,300 (6-core), 149,100 (24-core) and up

d

4.0 GHz (8): 55,100 (8-core), 204,300 (32-core) and up d

4.25 GHz (4) b : 162,100 (24-core), 330,000 (48-core) and up

d

Virtualization technologies

Optional PowerVM Virtualization Editions Standard, Enterprise

Maximum micro-partitions (10 per core) 254 e

IBM Chipkill ECC memory

Capacity 32 GB – 8 TB (1066 MHz DDR3)



Optional Memory CoD Trial, On/Off, CUoD

Active Memory™ Expansion Optional

Optional I/O drawers with PCI slots per system PCIe 12X I/O drawers Up to 32, each with 20 PCIe slots (max for #5873 is 31)

PCI-X DDR 12X I/O drawers Up to 32, each with 20 PCI-X slots

Maximum total 12X drawers per system 32

Hot-swappable disk/SSD bays Bays in CEC 0 (use I/O drawers) Bays in 24-inch 12X PCIe I/O drawer | Max. storage 26 SFF bays per #5803 (max 832) | 249 TB w/ 300 GB disks

Bays in 24-inch 12X PCI-X I/O drawer | Max. storage 16 SCSI bays per #5797/98 (max 512) | 153 TB w/ 300 GB disks

Bays in 19-in. EXP 12S SAS disk drawer | Max. storage (185 drawers max)

12 3.5-in. bays per #5886 (max 2220) | 999 TB w/ 450 GB disks

Bays in 19-in. EXP24 SCSI disk drawer | Max. storage (110 drawers max) – IBM i only

24 3.5-in. bays per #5786 (max 2640) | 792 TB w/ 300 GH disks

Maximum total bays per system 3052

Media bays Hot-swappable disk/SSD bays 0 in CEC (Use I/O drawer) Media bays in 24-inch system frame 2 (DVD & tape) using #5720 media drawer

Media bays in 19-inch racks 2 (DVD & tape) using 7214-1U2/1U3 (#0274)

Hot-plug GX++ adapter slots GX++ slots 4 per processor book (32 maximum per system)

Integrated ports HMC ports 4 USB ports, system ports, SPCN ports N/A

RAS capabilities Redundant, hot-plug power Y

Redundant, hot-plug cooling Y Node Add, Node Repair, Memory Upgrade Y (planned availability April 11, 2011) First failure data capture Y

Dynamic deallocation of system components Y Processor instruction retry / Alternate proc. recovery Y

Service processor Y

Other information AIX software tier Large

IBM i software tier P50

Warranty 1 year, 24x7 same-day response a – This value is the minimum number of cores that must be activated. The physical number of cores in a processor book is either 24 or 32. b – Each Power 795 4.0 GHz processor book has four sockets, with each socket having eight POWER7 processor cores. If run in optional TurboCore mode at 4.25 GHz, only half the cores in each socket are available, up to the number of cores activated. TurboCore mode requires that at least three processor books (96 cores) be installed. c – rPerf estimates shown for the Power 795 assume the LPAR size equal to the number of cores in one processor book (24-core for 3.7 GHz, 32-core for 4.0 GHz and 16-core for 4.25 GHz). d – CPW values are shown for minimum configurations – 6-core and 24-core 3.7 GHz, 8-core and 32-core 4.0 GHz, 24-core and 48-core 4.25 GHz (TurboCore). Use IBM Systems Workload Estimator to configure systems larger than those listed in this document. See http://www.ibm.com/systems/support/tools/estimator. e – IBM Statement of Direction to increase to 1000 the maximum number of Micro-Partitions on the Power 795 server.



IBM Power 710, 720, 730 and 740 Express Servers Q1: What performance benchmark results are available for the new Power Systems

Express servers?

A1: For additional detail about the new Power Systems Express servers, including performance

data, see:

Power 710 Express: http://www.ibm.com/systems/power/hardware/710/index.html



Power 740 Express; http://www.ibm.com/systems/power/hardware/740/index.html

For a summary of the characteristics of each of these servers, see the “At-a-glance” tables at

the end of this section of the Qs & As document.

Q2: Will these new Express systems have the ENERGY STAR rating?

A2: IBM has issued a Statement of Direction that we will target configurations of the Power 730

and Power 740 Express servers to be ENERGY STAR qualified.

Q3: Can you upgrade from a Power 710 to a Power 730 or from a Power 720 to a

Power 740?

A3: No. Upgrades from one model to another model within the POWER7 Express family are not

supported.

Q4: Can you start with one socket on a Power 730 or Power 740 and add a second socket

later?

A4: Only on the Power 740 Express can you start with one socket and add a second socket later.

The second processor module, which must be the same frequency and number of cores as the first

socket, requires CE installation. The Power 730 Express is a two-socket only offering.

Q5: Which systems support 12X external I/O drawers?

A5: Only the Power 720 and Power 740 Express support 12X external I/O drawers. Note that the

4-core Power 720 configuration does not support external I/O drawers, but 6-core and 8-core

Power 720 configurations do support external I/O drawers.

Q6: What performance benchmark results have been published for the new Power Systems

Express servers running the AIX OS? Are any of the results industry-leading?

A6: There are numerous one-socket and two-socket leadership benchmark results running the

AIX OS, including SPECint_rate2006, SPECfp_rate2006, SPECjbb2005and LINPACK. For

additional performance information, use the links in the answer to Question 1 above.

Q7: What performance benchmark results have been published for the new Power Systems

Express servers running the Linux OS? Are any of the results industry-leading?

A7: There are numerous one-socket and two-socket leadership benchmark results running the

Linux OS, including SPECint_rate2006, SPECfp_rate2006, SPECjbb2005 and SPEC ompM2001.

For additional performance information, use the links in the answer to Question 1 above.



Q8: What performance benchmark results have been published on the new Power Systems

Express servers running the IBM i OS? Are any of the results industry-leading?

A8: There will be a leadership Power 740 Express 2-socket, 16-core SAP performance - BI-

MXL result running the IBM i OS. For additional performance information, use the links in the

answer to Question 1 above.

Q9: What are the IBM estimates of relative performance (rPerf) for the new Express

servers?

A9: See answer to question 1 above. In addition, the At-a-glance tables below give the rPerf and

CPW ranges for each of the servers.

Q10: What are the options for Integrated Virtual Ethernet ports on these new Express

servers?

A10: Clients must choose from one of the following options, with FC 1824 being the default or

standard choice included with the system. These features are MES orderable.

Q11: What are the functions and client benefits provided with Integrated Virtual

Ethernet?

A11: Integrated Virtual Ethernet (IVE) is a shareable integrated Ethernet controller. IVE

provides a cost-effective Ethernet solution with performance equivalent to or better than provided

by dedicated Ethernet adapters. Using the HMC, the system administrator can allocate the IVE

logical ports to logical partitions (LPARs). The logical ports provide a standard Ethernet

interface to the host operating system in each LPAR. The logical IVE ports provide acceleration

and off-loading functions, including multiple receive queues

Q12: What is the GA date for the new Power Systems Express servers?

A12: General availability is September 17, 2010.

Q13: What operating systems and minimum technology levels are required for the new

Express Servers?

A13: The Power 710, 720, 730 and 740 Express servers are supported by the following operating

systems, or later:

• AIX 7.1

• AIX 6.1 with 6100-06 Technology Level

• AIX 5.3 with:

o 5300-12 Technology Level and Service Pack 1

o 5300-11 Technology Level and Service Pack 5 (available September 30, 2010)

o 5300-10 Technology Level and Service Pack 5 (available September 30, 2010)

• IBM i 7.1

• IBM i 6.1 with 6.1.1 machine code

• SUSE Linux Enterprise Server 11 Service Pack 1

• SUSE Linux Enterprise Server 10 Service Pack 3

• Red Hat Enterprise Linux AP 5 Update 5 (RHEL 5.5) for POWER

Description FC

Quad-port 1 Gbps Ethernet RJ45 (copper) 1824

Dual-port 10 Gbps Ethernet (fiber SR) 1825

Dual-port 10 Gbps Ethernet (copper TwinAx) 1826



If installing Virtual I/O Server (VIOS) software, VIOS 2.2, or later, is required.

There are unique considerations when running Java™ 1.4.2 on POWER7 servers. For best

exploitation of the outstanding performance capabilities and most recent improvements of

POWER7 technology, IBM recommends upgrading Java-based applications to Java 6 or Java 5

whenever possible. For more information refer to the following website:

http://www.ibm.com/developerworks/java/jdk/aix/service.html

Q14: Is the Integrated Virtualization Manager (IVM) supported on the new Express

Servers? A14: Yes. PowerVM partitions may be managed using built-in IVM software or optionally

through use of a Hardware Management Console (HMC). IVM is easy to use and helps reduce

the time and effort required to manage virtual devices, processors and partitions. An HMC is not

required for many functions, but if available, provides support for more virtual machines and a

overall higher level of systems management and control.

Q15: What is the maximum number of LPARs supported on the new Express Servers?

A15: The maximum number of virtual machines supported on the Power 710 and 720 is 80. The

maximum number on the Power 730 and 740 is 160 with HMC and 80 with IVM. For any

specific system, the maximum number of LPARs or Micro-Partitions is ten times the number of

processor cores in that configuration.

Q16: Is PowerVM supported on the new Express Servers?

A16: Yes. PowerVM Express, Standard and Enterprise Editions are supported on the new

Express servers.

Q17: Is redundant power an option on the new Express servers?

A17: Redundant hot-swap cooling fans are standard on all the new Express servers. Redundant

hot-swap power supplies are optional on the Power 710 and 720 Express, and standard on the

Power 730 and 740 Express.

Q18: Why type of PCIe slots are available in these server?

A18: All the slots are PCIe x8 slots. On the Power 710 and 730 Express, there are four

low-profile PCIe slots. The Power 720 and 740 Express have four standard full-high PCIe slots

and offer the option of four additional low-profile PCIe slots.

Q19: Are there any IBM Editions for the new Power Systems Express servers? If so, what

are they? What are the requirements?

A19: IBM Editions are available only as initial order. When the client orders a Power 710, 720,

730 or 740 Express server IBM Edition as defined below, they can qualify for half the processor

core activations on the initial configuration at no additional charge. IBM Editions are not

available in Greater China Group.

The features that determine whether the client is entitled to a processor activation at no additional

charge are:

• Total memory (based on the number of cores)



• Quantity and size of disk, SSD, Fibre Channel adapters or Fibre Channel over Ethernet

(FCoE) adapters shipped with the server

IBM Edition minimum memory definition details

A minimum of 2 GB memory per core is needed to qualify for the IBM Edition on the Power 710

and 720 Express, and a minimum of 4 GB per core on is required for the Power 730 and 740

Express. Note that on 6-core Power IBM Editions there is a 16 GB minimum memory

requirement on the Power 710 and 720 Express, and a 32 GB minimum memory requirement on

the Power 730 and 740 Express.

IBM Edition Minimum disk/SSD/Fibre Channel/FCoE definition details

A minimum of two disk drives, or two SSD drives, or two Fibre Channel adapters, or two FCoE

adapters is required. Clients must meet one of these disk/SSD/FC/FCoE criteria. Partial criteria

(for example, one disk drive and one Fibre Channel adapter) cannot be combined.

Configurations can be changed as needed and still qualify for processor entitlements at no

additional charge. However, selection of total memory or disk/SSD/Fibre Channel/FCoE adapter

quantities smaller than the totals defined as the minimums disqualifies the order as an IBM

Edition and the no-charge processor activations are then removed.

With an IBM Edition, clients can purchase an AIX, IBM i or Linux operating system license, or

they may choose to purchase the system with no operating system. The AIX, IBM i or Linux

operating system is processed via a feature number on the relevant OS order:

• AIX 5.3, 6.1, or 7.1

• IBM i 6.1.1 or IBM i 7.1

• SUSE Linux Enterprise Server or Red Hat Enterprise Linux

If the client chooses AIX 5.3, 6.1, or 7.1 for the primary operating system, they may also order

IBM i 6.1.1 or IBM i 7.1 and SUSE Linux Enterprise Server or Red Hat Enterprise Linux.

Similarly, if they choose an IBM i or Linux subscription as the primary operating system, they

may also order another operating system(s).

Q20: What is the standard warranty on these new Express servers?

A20: The standard warranty is 9 hours per day, Monday through Friday (excluding holidays),

next business day for three years at no additional cost. Warranty service upgrades and

maintenance are available.

Q21: Are these new Express servers customer setup (CSU) or CE setup?

A21: The new Express servers are customer setup offerings.

Q22: When is IBM announcing withdrawal of the Power 520 or Power 550?

A22: The Power 520 and Power 550 servers remain in market. IBM does not disclose future

product announcement or withdrawal plans.

Q23: What is the IBM SWG PVU (processor value unit) rating for the new Express

servers?

A23: The PVUs for the Power 710, 720, 730 and 740 Express servers is 70, which is identical to

the PVUs for one-socket and two-socket x86 processor-based servers.



Q24: What are the characteristics of the IBM Power 710 Express server?

A24:

IBM Power 710 Express server at a glance

General description

Machine type - Model 8231-E2B

System form factor 19" rack drawer (2U)


Processor packaging Planar

Number of sockets 1


3.0 GHz (4) | 4 3.7 GHz (6) | 6 3.55 GHz (8) | 8

EnergyScale Y L2 cache per core 256 KB

L3 cache per core 4 MB

Optional Processor Capacity on Demand (CoD) N/A

rPerf range 3.0 GHz (4) : 45.13 3.7 GHz (6) : 76.69 3.55 GHz (8) : 91.96

CPW range 3.0 GHz (4): 23,800 3.7 GHz (6): 40,900 3.55 GHz (8): 51,800


Optional PowerVM Virtualization Editions Express, Standard, Enterprise

Maximum micro-partitions (10 per core) 80

Integrated Virtual Ethernet

Choice of one: - Quad-port Gigabit (copper) - Dual-port 10 Gbps (copper) - Dual-port 10 Gbps (fibre)

IBM Chipkill ECC Memory

Capacity (install DIMMs in pairs) 8 GB – 64 GB (1066 MHz DDR3)

Optional Memory CoD N/A

Active Memory Expansion Optional

Internal storage Integrated storage controllers One SAS/SATA for disk/SSD/DVD

Backplane options (hot-swappable disk/SSD bays)

Choice of one: - 3 SFF disk/SSD bays + DVD + Tape

a

- 6 SFF disk/SSD bays + DVD a

- 6 SFF disk/SSD bays (w/ dual write cache RAID controller & external SAS port) + DVD

Media bays - Slimline DVD bay - Half-high bay (for optional tape drive or USB disk)

a

Maximum internal disk storage 1.8 TB

I/O adapter slots PCI Express (PCIe) 8X (Extended Error handling) 4 low profile

GX bus slots 1 GX++

Integrated capabilities

USB ports 3

HMC ports 2

System ports 2

SPCN ports N/A

Optional I/O drawers per system

EXP12S SAS disk drawer (#5886) b Up to 8, each with 12 3.5-in. hot-swap disk/SSD bays

Maximum bays in EXP 12S drawer | Max storage 96 bays | 43.2 TB w/ 450 GB disks PCIe 12X I/O drawers N/A PCI-X DDR 12X I/O drawers N/A



RAS capabilities

Redundant, hot-plug power Optional

Redundant, hot-plug cooling Y

First failure data capture Y

Dynamic deallocation of system components Y

Processor instruction retry / Alternate proc. recovery Y

Service processor Y

Other information Software tier Small

IBM i software tier P05 (4-core), P10 (6-core), P10 (8-core)

Warranty 3 year, 9x5 next business day a – Not supported on IBM i. b – Not supported on 4-core Power 710 configurations.


A25:

IBM Power 720 Express server at a glance

General description

Machine type - Model 8202-E4B

System form factor 19" rack drawer (4U) or tower


Processor packaging Planar

Number of sockets 1


3.0 GHz (4) | 4 3.0 GHz (6) | 6 3.0 GHz (8) | 8

EnergyScale Y


L3 cache per core 4 MB Optional Processor Capacity on Demand (CoD) N/A

rPerf range 3.0 GHz (4): 45.13 3.0 GHz (6): 65.52 3.0 GHz (8): 81.24

CPW range 3.0 GHz (4): 23,800 3.0 GHz (6): 34,900 3.0 GHz (8): 46,300







Capacity (install DIMMs in pairs) a

4-core: 8 GB – 64 GB (1066 MHz DDR3) 6- or 8-core: 8 GB – 128 GB (1066 MHz DDR3)





Choice of one:

- 6 SFF disk/SSD bays + DVD + Tape (opt. split backplane) b

- 8 SFF disk/SSD bays (w/ dual write cache RAID controller, split backplane & external SAS port) + DVD + Tape

Media bays - Slimline DVD bay - Half-high bay (for optional tape drive or USB disk

b)


I/O adapter slots PCI Express (PCIe) 8X (Extended Error Handling) 4 std + 4 low profile optional

GX bus slots 1 GX++ (not available if optional 4 PCIe slots used)


USB ports 3

HMC ports 2



System ports 2

SPCN ports 2

Optional I/O drawers per system EXP12S SAS disk drawer (#5886)

c Up to 28, each with 12 3.5-in. hot-swap disk/SSD bays

Maximum bays in EXP 12S drawers | Max storage 336 bays | 151 TB w/ 450 GB disks

PCIe 12X I/O drawers c

Up to 2 - #5802 w/ 10 PCIe slots & 18 SAS SFF disk/SSD bays - #5877 with 10 PCIe slots

Max bays in PCIe 12X I/O drawers (#5802) | Max storage 36 bays | 10.8 TB w/ 300 GB disks

PCI-X DDR 12X I/O drawers (#5796) c Up to 4, each with 6 PCI-X slots

RAS capabilities Redundant, hot-plug power Optional


First failure data capture Y Dynamic deallocation of system components Y


Service processor Y

Other information Software tier Small IBM i software tier P05 (4-core), P10 (6- or 8-core)

Warranty 3 year, 9x5 next business day a – In some configurations, memory DIMMs must be installed in quads. b – Not supported on IBM i. c – Not supported on 4-core Power 720 configuration.


A26:

IBM Power 730 Express server at a glance General description

Machine type - Model 8231-E2B System form factor 19" rack drawer (2U)

Processor Technology POWER7 Processor packaging Planar Number of sockets 2


3.0 GHz (4) | 8 3.7 GHz (4) | 8 3.7 GHz (6) | 12 3.55 GHz (8) | 16

EnergyScale Y


L3 cache per core 4 MB Optional Processor Capacity on Demand (CoD) N/A

rPerf range

3.0 GHz (4): 86.66 3.7 GHz (4): 101.62 3.7 GHz (6): 147.24 3.55 GHz (8): 176.57

CPW range

3.0 GHz (4): 44,600 3.7 GHz (4): 51,900 3.7 GHz (6): 77,200 3.55 GHz (8) : 97,700







Capacity (install DIMMs in pairs) 8 GB – 128 GB (1066 MHz DDR3)




Backplane options (hot-swappable disk/SSD bays) Choice of one: - 3 SFF disk/SSD bays + DVD + Tape

a

- 6 SFF disk/SSD bays + DVD a



- 6 SFF disk/SSD bays (w/ dual write cache RAID controller & external SAS port) + DVD

Media bays - Slimline DVD bay - Half-high bay (for optional tape drive or USB disk)

a


I/O adapter slots PCI Express (PCIe) 8X (Extended Error Handling) 4 low profile

GX bus slots 2 GX++

Integrated capabilities USB ports 3

HMC ports 2

System ports 2

SPCN ports N/A

Optional I/O drawers per system EXP12S SAS disk drawer (#5886) Up to 8, each with 12 3.5-in. hot-swap disk/SSD bays

Maximum bays in EXP 12S drawer | Max storage 96 bays | 43.2 TB w/ 450 GB disks

PCIe 12X I/O drawers N/A PCI-X DDR 12X I/O drawers N/A

RAS capabilities

Redundant, hot-plug power Y


First failure data capture Y

Dynamic deallocation of system components Y


Service processor Y

Other information Software tier Small IBM i software tier P20 Warranty 3 year, 9x5 next business day

a – Not supported on IBM i.


A27:

IBM Power 740 Express server at a glance General description

Machine type - Model 8205-E6B System form factor 19" rack drawer (4U) or tower


Processor packaging Planar Number of sockets 1 or 2


3.3 GHz (4) | 4 or 8 3.7 GHz (4) | 4 or 8 3.7 GHz (6) | 6 or 12

3.55 GHz (8) | 16

EnergyScale Y


L3 cache per core 4 MB

Optional Processor Capacity on Demand (CoD) N/A

rPerf range

3.3 GHz (4): 48.33 – 92.79 3.7 GHz (4): 52.93 – 101.62 3.7 GHz (6): 76.69 – 147.24

3.55 GHz (8): 176.57

CPW range

3.3 GHz (4): 25,500 – 47,800 3.7 GHz (4): 27,900 – 52,200 3.7 GHz (6): 41,600 – 77,200

3.55 GHz (8): 97,700









Capacity (install DIMMs in pairs) a 8 GB – 256 GB (1066 MHz DDR3)





Choice of one:

- 6 SFF disk/SSD bays + DVD + Tape (opt. split backplane) b

- 8 SFF disk/SSD bays (w/ dual write cache RAID controller, split backplane & external SAS port) + DVD + Tape

Media bays - Slimline DVD bay - Half-high bay (for optional tape drive or USB disk

b)


I/O adapter slots PCI Express (PCIe) 8X (Extended Error Handling) 4 std + 4 low profile optional GX bus slots 2 GX++ (1 GX++ if optional 4 PCIe slots used)


USB ports 3

HMC ports 2 System ports 2 SPCN ports 2

Optional I/O drawers per system EXP12S SAS disk drawer (#5886) Up to 28, each with 12 3.5-in. hot-swap disk/SSD bays

Maximum bays in EXP 12S drawers | Max storage 336 bays | 151 TB w/ 450 GB disks

PCIe 12X I/O drawers Up to 4 - #5802 w/ 10 PCIe slots & 18 SAS SFF disk/SSD bays - #5877 with 10 PCIe slots

Max bays in PCIe 12X I/O drawers (#5802) | Max storage 72 bays | 21.6 TB w/ 300 GB disks

PCI-X DDR 12X I/O drawers (#5796) Up to 8, each with 6 PCI-X slots

RAS capabilities Redundant, hot-plug power Y


First failure data capture Y Dynamic deallocation of system components Y


Service processor Y

Other information Software tier Small

IBM i software tier P20

Warranty 3 year, 9x5 next business day a – In some configurations, memory DIMMs must be installed in quads. b – Not supported on IBM i.



IBM Power 770 and 780 Server Topics

Q1: When will the large memory DIMMs (32 GB) for the Power 770 and 780 servers be

available?

A1: General availability for the 128 GB (4 x 32 GB DIMMs) memory feature #5602 for the

Power 770 and 780 servers has been improved with this announcement to September 17, 2010.

This memory feature enables configuration of these servers with the maximum memory capacity

of 2 TB. The minimum firmware level required is 7.2.0.

Q2: How many logical partitions (LPARs) are now supported on Power 770 and 780

servers?

A2: As of September 17, 2010, Power 770 and 780 servers will support up to 254 LPARs per

system. The minimum firmware level required is 7.2.0.

A previously issued IBM statement of direction, which states that IBM intends to increase the

maximum number of Micro-Partitions to 640 on Power 770 and 780 servers in the future,

continues in effect.



AIX 7.1 Q1: Is AIX 7 binary compatible with previous versions of AIX? A1: Yes, AIX 7 is backward binary compatible with AIX 6, AIX V5, and in some cases earlier

versions of AIX. Compatibility is subject to conditions and full details are available on

www.ibm.com/systems/power/software/aix/compatibility/.

Q2: Does IBM offer a ‘binary compatibility guarantee’? A2: Yes. IBM offers a binary compatibility guarantee that for well behaved 32-bit and 64 bit

programs created on AIX V5 and 32-bit programs created on AIX V4. Full details are available

at www.ibm.com/systems/power/software/aix/compatibility/.

Q3: Do I have to recompile my programs to use them on AIX 7? A3: No. As long as your program is well behaved and does not use non-portable programming

techniques, if the program ran on AIX 6 or AIX V5, it should run unchanged on AIX 7.

Q4: Why did the version change to ‘7’? A4: We changed the name to underscore the strategic importance of AIX to the IBM Corporation

and to show unity with POWER7.

Q5: Does AIX 7 support Linux Affinity? A5: Yes, all the Linux Affinity features of AIX 6 and AIX 5L are included.

Q6: What hardware is AIX 7 supported on? A6: AIX 7 supports IBM systems based on POWER4™, PowerPC® 970, POWER5 and

POWER6® processors.

Q7: How is AIX 7 priced? A7: AIX 7 is priced the same as AIX 6. There are three hardware size tiers (Small, Medium and

Large), and AIX 7 is available in three editions.

Q8: Is AIX 7 available in different editions?

A8: Yes, AIX 7 is available in Express, Standard and Enterprise Editions.

Q9: What is the difference between AIX 7 Express and AIX 7 Standard Editions? A9: AIX 7 (and AIX 6) Express Edition is limited to no more than 4 cores and 8GB of memory

per core (like AIX 6). Additionally, the Cluster Aware AIX features are not be available in AIX

Express Edition except for PowerHA SystemMirror. Finally, AIX Express Edition does not

include the AIX Profile Manager, but can be managed by AIX Profile Manager.

Q10: Can I upgrade from AIX 5.3 to AIX 7 Enterprise Edition? A10: Yes, but you have to do it in two steps – first upgrade to AIX 7 Standard Edition then do an

MES upgrade to AIX 7 Enterprise Edition.

Q11: Do I have to reinstall AIX 7 if I move from Express to Standard Edition? A11: No, you can change the edition by ordering the relevant upgrade feature and then using the

chedition command.



Q12: What does AIX 7 Enterprise Edition include? A12: AIX 7 and AIX 6 Enterprise Editions include:

• Either AIX 6 or AIX 7 Standard Edition

• plus Workload Partitions Manager

• plus Systems Director Enterprise Edition (which includes Active Energy Manager,

VMControl Enterprise, and Tivoli Monitoring and TADDM along with other

components)

Q13: Does the use of the 64-bit kernel mean that I will have to get new 64-bit versions of my

application? A13: No. AIX has long supported execution of 32-bit and 64-bit applications by the 64-bit

kernel. Existing 32-bit applications should run on AIX 7 without change or recompilation.

Q14: Where can I find more information about AIX 7? A14: There are several Web sites with information about AIX 7, including www.ibm.com/AIX,

www.redbooks.ibm.com and www.ibm.com/developerworks. In addition, the AIX 7 publications

are available on the Web.

Q15: Will AIX 7 provide a significant performance boost over AIX V5.3? A15: Yes. On POWER7 servers, AIX 7 can run in POWER7 mode which provides for

Simultaneous Multi-threading 4 (SMT4). The combination of SMT4 and memory affinity

exploitation may result in significant performance improvement, compared to AIX 5.3.

Q16: Will AIX 7 provide a significant performance boost over AIX 6? A16: AIX 7 should have similar performance as AIX 6 when run on the same hardware.

Q17: Do I have to run AIX 7 on POWER7? A17: No. Systems based on POWER7 processors are supported by AIX V5.3 and AIX 6, as well

as AIX 7.

Q18: How much does it cost to upgrade to AIX 7? A18: An upgrade to AIX 7 from previous versions of AIX is included in the Software

Maintenance Agreement (SWMA). Clients that have SWMA on their current AIX systems can

upgrade for no additional change.

Q19: Can I upgrade to AIX 7 from previous releases of AIX? A19: Yes. You can upgrade directly to AIX 7 from earlier versions of AIX V5 and AIX 6

without having to upgrade to intermediate releases of AIX. The upgrade process uses proven

techniques such as the alternate disk installations, NIM and migration installations.

Q20: What features of AIX 7 are available only on POWER7 processor-based systems? A20: Most features of AIX 7 are available on all supported systems. A few features, such as

SMT4 and 256-core/1024-thread partitions, are available only when using AIX 7 on systems

based on POWER7 processors.

Q21: What POWER7 features are available only for AIX 7? A21: Support for 256 cores and 1024 threads in a single partition is available only with AIX 7.



Q22: Will AIX 6 support more than 64 cores / 256 threads in a single partition? A22: No.

Q23: Do I have to run AIX Enterprise Edition to get 256-core/1024-thread support? A23: No, AIX 7 Standard and Enterprise Editions support 256 cores/1024 threads.

Q24: Have many clients taken advantage of the AIX Open Beta? A24: Yes. Several thousand clients and ISVs have participated in the AIX 7 Open Beta

Q25: What are Workload Partitions (WPARs)? A25: Workload Partitions (WPARs) is a software based virtualization technique that was first

introduced in AIX 6. It allows a client to run multiple independent workloads inside of a single

AIX instance (called a “global instance”). Each WPAR can be set up to “look” like a separate

system, with its own root administrator and separate security context (users and groups). Each

WPAR shares the single AIX kernel.

Q26: What are the key benefits of WPARs? A26: WPARs provide an alternative that enables more efficient administration by reducing the

number of AIX operating system instances that have to be managed.

AIX 5.2 Workload Partitions for AIX 7 Q1: Can I run AIX 5.2 in an LPAR on POWER7? A1: No. AIX 5.2 is not supported running in an LPAR on POWER7 servers.

Q2: Can I run AIX 5.2 in a Workload Partition (WPAR)? A2: Yes, if you purchase the AIX 5.2 Workload Partitions for AIX 7 product.

Q3: How do you get AIX 5.2 in a WPAR? A3: After purchasing and installing the AIX 5.2 WPAR for AIX 7 product into an LPAR, you

then back up your existing legacy AIX 5.2 system and restore it into the WPAR with a new flag

on the mkwpar command or though the WPAR Manager.

Q4: What are the prerequisites for running AIX 5.2 in a WPAR? A4: The prerequisites are:

• The AIX 5.2 WPAR for AIX 7 product

• AIX 7 to host the WPAR

• Must run on a POWER7 server

Q5: Why would I run AIX 5.2 in a WPAR? A5: In most cases, clients that are still running AIX 5.2 are running it on very old POWER4 or

earlier generations of hardware. While those systems were great in their time, they have very

poor performance compared to POWER7 processor-based systems, and they use a lot more

energy per unit of work. Consolidating these workloads on POWER7 servers in AIX 5.2 WPARs

can potentially lower energy costs, improve performance, free up floor space and reduce

hardware maintenance charges for obsolete hardware.

Q6: Do I have to purchase AIX 5.2 extended support to run AIX 5.2 in a WPAR? A6: No. The Software Maintenance Agreement for the AIX 5.2 WPARs for AIX 7 product

includes phone and limited new fix support for AIX 5.2 running in a WPAR.



Q7: Can I run AIX 5.2 WPARs on POWER6 processor-based servers? A7: No. AIX 5.2 WPARs for AIX 7 is supported only on POWER7 servers.

Q8: Can I run AIX 5.1 or AIX 4.3 inside of a WPAR? A8: No. AIX versions other than AIX 5.2 will not run inside of a WPAR with this product.

Q9: Why would I run AIX 5.2 inside of a WPAR? A9: Many clients have secondary workloads still running on AIX 5.2 on older POWER hardware.

This new capability provides an easy way to consolidate those old environments on POWER7

systems.

Q10: Do I have to buy AIX 7 licenses to run AIX 5.2 WPARs? A10: Yes. AIX 5.2 WPARs runs on top of AIX 7. You need to purchase both AIX 7 and the

AIX 5.2 WPARs products.

Q11: Do I have to run AIX 5.2 WPARs on AIX Enterprise Edition? A11: No. AIX 5.2 WPARs will run on AIX 7 Express, Standard or Enterprise Editions.

Q12: Is there any instruction translation involved in running AIX 5.2 in a WPAR? A12: No, there are just a few system calls that return the AIX 5.2 structure when called from

AIX 5.2 inside a WPAR. All processes are running at full native POWER7 speed.

Q13: What about support for AIX 5.2? A13: This offering includes limited support for AIX 5.2 running in this environment, including

phone support and potentially new fixes for critical security problems.

Q14: Does the AIX 5.2 support require purchase of standard extended support? A14: No. Support for AIX 5.2 inside a WPAR is included in the SWMA for the AIX 5.2

WPARs for AIX 7 product.

Q15: If customer has an earlier AIX 5.2 and wants to move to Technology Level 10 (TL10)

under AIX 5.2 WPAR, will we offer AIX download from Entitled Support SW? A15: Entitled Support Software (ESS) does not contain AIX 5.2 images any longer because

AIX 5.2 has been withdrawn from marketing. Clients can download Technology Level updates

from Fix Central to update their AIX 5.2 systems to the supported level.

Q16: Does the AIX 5.2 WPAR for AIX 7 product include AIX 5.2 installation CDs? A16: No. This product is intended for customers that are already running AIX 5.2 on legacy

hardware. A mksysb backup of the legacy system is the only way to create an AIX 5.2 WPAR.

Q17: What about independent software vendor (ISV) support? Will ISVs support running

inside of an AIX 5.2 WPAR? A17: The question of ISV support has to be answered by each ISV. We have found that, in the

majority of cases of clients running AIX 5.2 on legacy hardware, the ISV applications are also

very old and out of support.



Q18: Does AIX 5.2 running in a WPAR exploit hardware cryptographic adapters? A18: No. AIX 5.2 does not include the cryptographic framework necessary to exploit the

cryptographic adapter support included with AIX 7.

Q19: Can I run HACMP™ / PowerHA SystemMirror inside of a WPAR running AIX 5.2? A19: No. You will need to use PowerHA SystemMirror in the global instance to manage

availability for AIX 5.2 running in a WPAR.

Q20: Can I have SAN devices inside of an AIX 5.2 WPAR? A20: Yes.

Q21: Can I use virtual SCSI devices in AIX 5.2 inside of a WPAR? A21: Yes.

Q22: What are some of the difference when running AIX 5.2 in a WPAR compared to a

legacy environment? A22: There are the benefits of POWER7 performance, including support for SMT4 and mobility.

But not all devices are available when running in a WPAR.

Q23: Is AIX 5.2 in a WPAR running the AIX 5.2 kernel? A23: No. The kernel used is the AIX 7 kernel. This has several positive implications, including

support for SMT4 and mobility and support for all POWER7 features.

Q24: Are all commands and libraries from AIX 5.2 the same when running in a WPAR? A24: Due to the relationship with the kernel, about 100 performance related commands such as

vmstat are overlaid with the AIX 7 version of the command when creating and AIX 5.2 WPAR.

Q25: What kind of devices cannot be used inside of AIX 5.2 in a WPAR? A25: The WPAR I/O support model does not support directly attached network devices. All

network I/O is aliased over a physical adapter owned by the global instance. Other types of I/O

devices not currently supported include multi-port asynchronous adapters, token ring and any

device not supported by AIX 7.

Q26: Is mobility supported for AIX 5.2 WPARs? A26: Yes, both Live Partition Mobility (relocating the entire LPAR with the WPAR inside) and

Live Application Mobility (relocating the WPAR to another AIX 7 system) are supported.

Q27: Can different AIX 5.2 WPARs have different performance tuning? A27: Performance tuning for WPARs occurs at the global instance level, so all WPARs inherit

the tuning from the global instance. If different WPARs require different tuning, then they should

be placed inside separate LPARs.

Q28: What filesystems can I use in an AIX 5.2 WPAR? A28: WPARs support NFS and JFS2 filesystems. If the original AIX 5.2 system used JFS

filesystems, those filesystems are automatically converted to JFS2.



IBM Service Delivery Manager & IBM CloudBurst Q1: What is the primary focus of today’s Statement of Direction for IBM Service Delivery

Manager?

A1: IBM is announcing its intention to offer clients a choice in supported hardware platforms for

the deployment of the IBM Service Delivery Manager software. It is planned to extend the

supported platforms offered for this solution to IBM Power Systems.

IBM Power Systems, with its differentiated virtualization, provide enterprise quality of service

capabilities including performance, scalability, reliability, and security, enabling clients to move

mission-critical workloads to the cloud. Now, clients can plan to couple the industrial strength

virtualization of Power Systems with Tivoli service management capabilities such as seamless

self-service and automated provisioning, to deliver an integrated cloud environment. The IBM

Service Delivery Manager offering for Power Systems is planned to provide a set of pre-

integrated and pre-configured virtual appliances that simplify and accelerate the deployment of

the cloud management infrastructure and help to provide flexibility in the selection of cloud

resources to be provisioned.

Building on IBM’s previous announcement of its intention to extend the platforms offered for the

IBM CloudBurst™ integrated solution to include IBM Power Systems. Today’s statement of

direction reinforces IBM’s commitment to provide clients with a wide range of integrated, pre-

tested hardware and software solutions for implementing a cloud infrastructure. These expanded

offerings will allow clients to leverage all the benefits of Power Systems for both the cloud

management infrastructure and cloud computing environment.

Q2: What is IBM Service Delivery Manager?

A2: IBM Service Delivery Manager is a pre-integrated software stack, deployed as a set of

virtual images that automate IT service deployment and provide resource monitoring, cost

management, and availability of services in a cloud. It enables the data center to accelerate the

creation of service platforms for a wide spectrum of workload types with a high degree of

integration, flexibility, and resource optimization.

IBM Service Delivery Manager is for enterprise customers who want to get started with a private

cloud computing model. The product enables you to rapidly implement a complete software

solution for service management automation in a virtual data center environment, which in turn

can help your organization move towards a more dynamic infrastructure.

IBM Service Delivery Manager is a single solution that provides the necessary software

components to rapidly implement a cloud computing environment. Cloud computing is a services

acquisition and delivery model for IT resources, which can help improve business performance

and control the costs of delivering IT resources to an organization.

Q3: When will IBM Service Delivery Manager be available?

A3: IBM Service Delivery Manager v7.2.1 supporting deployment on x86 announced today with

a planned availability date of August 27, 2010. IBM intends to deliver the IBM Service Delivery

Manager offering supporting deployment on Power Systems at a future date.

Q4: What are the core capabilities of IBM Service Delivery Manager?



A4: IBM Service Delivery Manager provides preinstalled capabilities essential to a cloud service

delivery model, including:

• A self-service portal interface for reservation of computer, storage, and networking

resources, including virtualized resources

• Automated provisioning and de-provisioning of resources

• Prepackaged automation templates and workflows for most common resource types

• Service management for cloud computing

• Real time monitoring for elasticity

Q5: What are the potential benefits of IBM Service Delivery Manager?

A5: Benefits of IBM Service Delivery Manger include:

• Reduction in operational and capital expenditures

• Enhanced productivity - the ability to innovate more with fewer resources

• Decreased time-to-market for business features that increase competitiveness

• Standardized and consolidated IT services that drive improved resource utilization

• Increased resiliency to market demands

• Improved quality of service for IT consumers

Q6: What are the hardware requirements for installing IBM Service Delivery Manager?

A6: IBM Service Delivery Manager is designed to operate on an x86 server platform. Support

for Power Systems is planned for a future date. For a comprehensive listing of all hardware

requirements refer to the “IBM Service Delivery Manager Installation and Configuration Guide”.

Q7: What ordering system do I use to order IBM Service Delivery Manager?

A7: IBM Service Delivery Manager is only orderable via the PassPort Advantage (PPA) system.

Q8: Is IBM Service Delivery Manager available to the business partner channels?

A8: Yes

Q9: Where can I get more information about IBM Service Delivery Manager?

A9: IBM Service Delivery Manager information is being combined with the IBM CloudBurst

sales kit and can be found at:

http://w3-103.ibm.com/software/xl/portal/content?synKey=X763620L98422Y00#overview

The Business Partner sales kit is pending availability.

Q10: How does IBM Service Delivery Manager differ from the IBM CloudBurst offering?

A10: IBM CloudBurst and IBM Service Delivery Manager offer similar software features,

functions, and benefits with one of the major differences being the fact that IBM Service Delivery

Manager is a software only solution that can be deployed on existing hardware.

IBM CloudBurst includes similar software capabilities with the addition of energy management

and storage virtualization that are factory installed and pre-configured on defined set of IBM

hardware, storage and network components. IBM CloudBurst can help reduce the amount of

guesswork behind system configuration needs including:

• Operating System requirements

• Hardware Platform requirements



• Multiple software application installations including system compatibility and coexistence

requirements

IBM Service Delivery Manager may appeal to clients who already have invested in hardware

(including servers, storage and networks) and wish to purchase a cloud service delivery software

stack separate from the hardware. For this reason, IBM Service Delivery Manager is being

offered as an orderable set of software virtual images.

Q11: What is IBM CloudBurst?

A11: The IBM CloudBurst product is designed from IBM client cloud implementation

experiences and integrates the service management software system with servers, storage, and

Quickstart services to enable a private cloud in the client’s IT environment. It takes the

guesswork out of establishing a private cloud by pre-installing and configuring the necessary

software on the hardware and leveraging services for customization to your environment. All the

client must do is install their applications and start leveraging the benefits of cloud computing,

such as virtualization, flexibility, scalability and a self service portal for provisioning new

services.

The current release of CloudBurst has been optimized for application development and test

environments, but can also host other workloads. It allows the data center to accelerate the

creation of services for a variety of workloads with a high degree of flexibility and resource

optimization. IBM CloudBurst provides an alternative from traditional IT infrastructure. It is

easy for customer to acquire, implement and use. Here are some of the highlights:

• Single price

• Single delivery

• Single install

• Quickstart services

• Single support

• Single phone number to call for all support issues

• Single vendor interface for all hardware and software

Q12: When will IBM CloudBurst be available?

A12: IBM CloudBurst v2.1 on the IBM BladeCenter platform is available today. IBM intends to

deliver the IBM Cloudburst offering with Power Systems at a future date.

Q13: What are some benefits clients will see with IBM CloudBurst?

A13: Benefits include:

• Lowers total cost of ownership by 50 percent or more with reduced labor for configuration,

operations, management and monitoring of the test environment

• Reduced test provisioning cycle times from weeks to minutes

• Can eliminate up to 30 percent of all defects that come from faulty configurations

• Improved business innovation

• Enables real time sharing of information & data streams

• Provides open, elastic IT environments

• Supports globalization and globally available resources

Q14: How do Business Partners participate in CloudBurst solutions?



A14: We expect Business Partners to contribute significantly to the success of IBM CloudBurst

offerings long term and have received extremely positive feedback from them about our solutions.

Currently, IBM Business Partners do not have a role in the sale or implementation of IBM

Cloudburst. At this time we are encouraging Business Partners with both IBM Power Systems

and IBM Tivoli Automation and Provisioning experience to:

• Learn more about IBM CloudBurst by visiting the website and attending two informative

webcasts

• Build technical skills by becoming involved in beta technical enablement programs

• Watch IBM PartnerWorld® for the latest demand generation materials

In the future, we are planning to provide Business Partners with new incentives to sell the

Cloudburst offerings as well as expanded opportunities to drive implementation and

customization services.

Q15: Where can I get more information about IBM CloudBurst?

A15: The IBM CloudBurst sales kit can be found here: http://w3-

103.ibm.com/software/xl/portal/content?synKey=X763620L98422Y00#overview

External landing page: http://www-01.ibm.com/software/tivoli/products/cloudburst/

Product documentation:

https://www.ibm.com/developerworks/wikis/display/tivolidoccentral/IBM+CloudBurst

Q16: Is CloudBurst the only option for creating a private cloud? Can my customer do

something now on Power Systems?

A16: IBM CloudBurst is one choice for creating a private cloud. It is an integrated solution that

can give customers a quick start for private cloud implementations. It is not the only option. We

have been creating private clouds leveraging all IBM platforms for the past two years using an

offering now called the IBM Smart Business Test and Development Cloud. Private clouds can

also be built by customers using the IBM products that are found pre-packaged in CloudBurst,

including Tivoli Service Automation Manager and Power Systems. See the following whitepaper

for the advantage of creating a private cloud on Power Systems https://w3-

03.ibm.com/sales/competition/compdlib.nsf/weball/8870201BFB11A789852576CD005C7808?O

penDocument

Q17: How does IBM CloudBurst differ from IBM WebSphere® Cloudburst?

A17: These products are complementary, and together they accelerate payback in a private cloud

environment. The IBM WebSphere CloudBurst Appliance helps bring together deep expertise in

building and managing application environments, while IBM CloudBurst creates broad cloud

management capabilities, along with the necessary hardware.

WebSphere CloudBurst Appliance is a hardware appliance that provides access to software

virtual images and patterns that can be used as-is, or easily customized, securely deployed,

managed, and maintained in a private cloud. The first hardware appliance of its kind, the

WebSphere CloudBurst Appliance stores and secures WebSphere Application Server Hypervisor

Edition images and patterns to be dispensed into a cloud. It helps customers easily and quickly

develop, test and deploy business applications, ending the use of manual, complex or time-

intensive processes associated with creating application environments.



IBM CloudBurst is a complete IBM Service Management package of hardware, software and

services, which simplifies your cloud computing acquisition and deployment. Currently built on

the IBM BladeCenter platform, but with plans to expand to the IBM Power Systems platform,

IBM CloudBurst provides pre-installed, fully integrated service management capabilities across

hardware, middleware and applications.

WebSphere CloudBurst Appliance helps developers and operations personnel establish and

deploy software images and patterns into a cloud environment, but does not include the

infrastructure for that cloud environment. IBM CloudBurst offers a ready-made cloud

environment into which these images and patterns can be deployed, and is designed to be used by

an organization that doesn’t want to create a cloud environment using existing assets.

IBM Statements of Direction IBM is committed to enhancing its clients' investments in IBM Power System servers and related

software. Based on this commitment, IBM plans to provide the following future enhancements.

• IBM Service Delivery Manager intends to offer a choice in supported hardware platforms

for the deployment of the IBM Service Delivery Manager software. It is planned to

extend the supported platforms offered for this solution to IBM Power Systems.

• IBM plans to support the following AIX technology levels on the Power 795 server and

on Power 710, 720, 730 and 740 Express servers:

� AIX 6.1 with the 6100-05 Technology Level and Service Pack 3, or later



• IBM plans to enhance its Power Systems Enterprise Class SSD solutions with technology

designed to continue to provide significant improvements in performance and storage

density over time. IBM plans for these IBM Power Systems enhancements to include

both SAS bay based and PCIe based SSD product offerings that will leverage IBM's

investments in its SSD optimized Enterprise Class RAID Storage Controllers.

• IBM plans to support installations of selected Power 795 configurations in non-raised

floor environments. IBM also plans to deliver options to route communications cables

overhead through the top of a rack. Availability is planned for first half of 2011.

• IBM plans to provide capability for Dynamic Power Saver modes to be enabled on

dedicated operative system partitions via Active Energy Manager. This allows the

flexibility to have some partitions running in a power saving mode, others in a potentially

performance enhancing mode, and others running with the nominal system settings.

• IBM intends to enhance PowerHA SystemMirror Enterprise Edition to exploit AIX cluster

aware technologies.

• IBM plans to support the Broadcom 2/4-port Ethernet Expansion Card (CFFh)

(FC 8291 on BladeCenter JS23, JS43, PS700, PS701 and PS702 Express) with the

following AIX technology levels:

� AIX V6.1 with the 6100-05 Technology Level and Service Pack 3, or later





• IBM plans to support these features:

� 177GB SSD Module with eMLC (FC 1995)

� PCIe RAID & SSD SAS Adapter 3Gb(FC 2054)

� PCIe RAID & SSD SAS Adapter 3Gb with Blind Swap Cassette (FC 2055), and

� Package of five FC 2055 & twenty FC 1995 (FC 4367)

with the following AIX technology levels:




• IBM intends for PowerVM Lx86 to support the next major release of Red Hat Enterprise

Linux on IBM Power Systems.

• Red Hat intends for the upcoming release of Red Hat Enterprise Linux 6 to support the

latest POWER7 models – Power 795 and Power 710, 720, 730 and 740 Express.

• IBM is pursuing Energy Star computer server qualification for certain configurations of

the Power 730 and 740 Express Servers.

• IBM intends to offer a DC power supply on the Power 750 Express in 4Q/2010.

IBM also intends to complete conformance testing of the Power 750 Express and the

BladeCenter PS700, PS701 and PS702 Express to Network Equipment-Building System

(NEBS) Level 3 and ETSI in 4Q/2010.

• IBM plans to add Linux on Power development tools to the Rational Developer for Power

family in the future, extending the benefits of having an integrated Eclipse development

environment to teams developing applications for Linux on Power.

All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and

represent goals and objectives only. Any reliance on these Statements of Direction is at the relying party's sole risk

and will not create liability or obligation for IBM.

Information on new products is intended to outline our general product direction and it should not be relied on in

making a purchasing decision. The information on the new product is for informational purposes only and may not be

incorporated into any contract. The information on the new product is not a commitment, promise, or legal obligation

to deliver any material, code or functionality. The development, release, and timing of any features or functionality

described for our products remains at our sole discretion.



Footnotes

SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPEC OMP, SPECviewperf, SPECapc, SPEChpc, SPECjvm, SPECmail, SPECimap and SPECsfs are trademarks of the Standard Performance Evaluation Corp (SPEC). Other company, product, and service names may be trademarks or service marks of others. The IBM home page on the Internet can be found at http://www.ibm.com. The IBM Power Systems home page on the Internet can be found at http://www.ibm.com/systems/power. More information about AIX can be found at http://www.ibm.com/aix

© Copyright IBM Corporation 2010

IBM Corporation Marketing Communications Systems and Technology Group Route 100 Somers, New York 10589 Produced in the United States of America August 2010 All Rights Reserved This document was developed for products and/or services offered in the United States. IBM may not offer the products, features, or services discussed in this document in other countries. The information may be subject to change without notice. Consult your local IBM business contact for information on the products, features and services available in your area. This equipment is subject to FCC rules. It will comply with the appropriate FCC rules before final delivery to the buyer. IBM hardware products are manufactured from new parts, or new and used parts. Regardless, our warranty terms apply. All statements regarding IBM’s future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. Information concerning non-IBM products was obtained from the suppliers of these products. Questions on the capabilities of the non-IBM products should be addressed with the suppliers. All performance information was determined in a controlled environment. Actual results may vary. Performance information is provided “AS IS” and no warranties or guarantees are expressed or implied by IBM. IBM, the IBM logo, Active Memory, AIX, BladeCenter, CloudBurst, DB2, EnergyScale, HACMP, InfoSphere, PartnerWorld, Power, POWER4, POWER5, POWER5+, POWER6, POWER7, PowerHA, PowerPC, PowerVM, Power Systems, Rational, Tivoli and WebSphere are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries or both. A full list of U.S. trademarks owned by IBM may be found at: http://www.ibm.com/legal/copytrade.shtml. The Power Architecture and Power.org wordmarks and the Power and Power.org logos and related marks are trademarks and service marks licensed by Power.org. UNIX is a registered trademark of The Open Group in the United States and other countries. Linux is a trademark of Linus Torvalds in the United States, other countries or both.

Aug17 2010 annc faq

Documents

ibm announcements

ibm power systems qs

management ibm

todays ibm

popular power

power family

pretested power systems

new power systems servers