Transcript AIX Editions

IBM Power Systems
AIX 6 Excellence
Ravi Singh
IBM Power Systems
[email protected]
1
© 2009 IBM Corporation
IBM Power Systems
AIX 6
2
© 2009 IBM Corporation
IBM Power Systems
Characteristics of a good OS - AIX
 Availability
Power Systems with AIX delivers 99.997% uptime
 Virtualization support
Optimized and integrated with POWER and PowerVM
 Security
Compliant with several security certifications and standards
 Performance
Industry leading benchmarks
 Roadmap
Strong and stable roadmap consistent with new releases
 ISV Adoption
More than 11,000 applications available
 Industry standard Features
UNIX, Security and Open standards
 Innovative and unique features
Features available only in AIX
3
© 2009 IBM Corporation
IBM Power Systems
20+ Years of AIX Progress
1986-1992 1994-1996 1997-1999 2001
2002
2004-2007 2008-09
AIX/6000
AIX V2 & V3
Establishment in
the market:
- RISC Support
- UNIX credibility
- Open Sys. Stds..
- Dynamic Kernel
- JFS and LVM
- SMIT
AIX V3.2.5
Maturity:
- Stability
- Quality
AIX V4.1/4.2
SMP Scalability:
- POWERPC spt.
- 4-8 way SMP
- Kernel Threads
- Client/Server pkg
- NFS V3
- CDE
- UNIX95 branded
- NIM
- > 2GB filesystems
-HACMP Clustering
- POSIX 1003.1,
1003.2, XPG4
- Runtime Linking
- Java 1.1.2
AIX V4.3
Higher levels of
scalability:
- 24-way SMP
- 64-bit HW support
- 96 GB memory
- UNIX98 branded
- TCP/IP V6
- IPsec
- Web Sys. Mgr.
- LDAP Dir. Server.
- Workload Mgr
- Java JDT/JIT
- Direct I/O
- Alt. Disk Install
- Exp/Bonus CDs
AIX V5.1
Industry Leading
Performance:
- POWER4 support
- Static LPAR
- Linux Affinity
- New 64bit kernel
- 32-way SMP
- 256GB mem
- JFS2
- Networking enh.
- Java 2 support
- Dynamic CPU
Deallocation
- Cluster Mgt (CSM)
- GRID Toolkit
AIX V5.3
AIX V5.2
Flexible Resource
Management:
- POWER4+ spt.
- Dynamic LPAR
- Dynamic CUoD
- Dyn. CPU Sparing
- 512GB mem
- 16 TB filesystems
- UNIX03 branded
- Concurrent I/O
- MultiPath I/O
- Mobile IP V6
- System UE Gard
- Flex LDAP Client
- XSSO PAM spt
AIX V6.1
Advanced
Virtualization:
- POWER5 spt.
- 64-way SMP
- SMT
- MicroPartitions™
- Virt I/O Server
- Partition Load Mgr
- NFS Version 4
- Adv. Accounting
- Scaleable VG
- JFS2 Shrink
- SUMA
- SW RAS features
- POSIX Realtime
PowerVM
Virtualization:
- POWER6 support
- Workload Partitions
- Application Mobility
- Continuous Avail.
- Storage Keys
- Dynamic tracing
- Software FFDC
- Recovery Rtns
- Concurrent MX
- Trusted AIX
- RBAC
- Encrypting JFS2
- AIX Security Expert
- Compliance Expert
- AIX Runtime Expert
Open Systems Distributed Network Centric
Workstations Client-Server Computing
Uni-processor
4
4-8 way SMP
24-way SMP
e-Business
Computing
32-way SMP
On Demand
Business
32-way SMP
POWER of 6
64/way SMP
© 2009 IBM Corporation
IBM Power Systems
Unsurpassed Reliability
According to a recent Yankee Group study* of
400 Windows, Linux and UNIX users, AIX
was the most reliable server operating
system:
Hours of downtime per year*
“IBM’s AIX achieved the highest
level of reliability, with corporate
enterprises reporting and
average of only 36 minutes of
downtime per server in a 12month period”
9
8
7
6
5
4
3
2
1
0
AIX
HP-UX
Solaris
Windows
* Source: “Unix, Linux Uptime and Reliability Increase; Patch Management Woes Plague Windows” © 2008 Yankee Group Research, Inc. All rights reserved
5
© 2009 IBM Corporation
IBM Power Systems
Power Systems with AIX deliver 99.997% uptime
- 54% of IT executives and managers say that they require 99.99% or better availability
for their applications
Power Systems with AIX delivers the best RAS
of UNIX, Linux, and Windows
1. Availability: The least amount of downtime
 15 minutes a year
 2.3 times better than the closest UNIX
competitor
 more than 10X better than Windows
2. Reliability: The fewest unscheduled outages
 less than one outage per year
3. Serviceability: The fastest patch time
 11 minutes to apply a patch
Minutes of downtime per year
180
120
60
0
AIX / Power
HP-UX /
PA_RISC
HP-UX /
Integrity
x86 - Windows
Source: ITIC 2009 Global Server Hardware & Server OS Reliability Survey Results, July 7, 2009
6
© 2009 IBM Corporation
IBM Power Systems
IBM AIX on the Power series leads all vendors for both server
hardware and server OS reliability
Ref: ITIC 2009 Global Server Hardware and Server OS Reliability Survey
7
© 2009 IBM Corporation
IBM Power Systems
New Security Capabilities
New feature
Benefit to customer
Role Based Access Control
 Tighter security by reducing the number of root users
 Improved IT efficiency by allowing administrative tasks to be
delegated to non-root users
Encrypting Filesystem
 Critical data is safeguarded from loss, even from root
 Improved IT efficiency though automated key management
AIX Security Expert
 Tightened security by federating security settings across the
enterprise
 Improved IT efficiency by reducing administrative effort of security
management
Secure by Default
 Tighter security by requiring all services to be explicitly enabled
 Reduced administrative effort when used with AIX Security Expert
Trusted Execution
 Stronger resistance to hacks and penetrations by validated execution
Trusted AIX and Planned
CAPP/RBACPP/LSPP/EAL4+
Certifications
 Highest level of compartmentalized security for demanding workloads
 Improved IT efficiency though automated key management
8
© 2009 IBM Corporation
IBM Power Systems
New Continuous Availability Capabilities
New feature
Benefit to customer
POWER6™ Storage Keys
 Improved availability by reducing system and application outage
caused by storage overlays
 Reduced time to repair by eliminating a whole class of intermittent
problem
Concurrent AIX updates
 Eliminate some planned system outages because some kernel fixes
to be installed without rebooting
 Potentially improved security by enabling administrators to put on
some critical security fixes without waiting for an outage window
Live Application Mobility
& Live Partition Mobility
 Greater application availability though moving applications off of a
system that is going to be taken down
 Greater application availability by allowing workloads to be moved
off of overloaded servers with impacting the end user
probevue Dynamic Tracing
 Greater application reliability through improved performance and
fault analysis
Functional Recovery Routines
 Improved operating system availability through self healing
programming
Mainframe inspired availability
features
 Greater reliability though easier problem determination and
resolution
9
© 2009 IBM Corporation
IBM Power Systems
New Virtualization Capabilities
New capability
How capability addresses client needs
Workload Partitions
 Improved IT efficiency through reduced number of AIX instances
than need to be managed
 Greater flexibility by simplifying consolidation and virtualization
 Improved manageability by allowing administrators to group all
application resources and manage as a whole
 New workload opportunities though consolidation of hundreds of
workloads on a single AIX instance
Live Application Mobility
& Live Partition Mobility
 Increased application availability by moving applications to avoid
outages
 Improved IT efficiency though relocation of workloads to lower used
server
 Increased energy efficiency by consolidating workloads on fewer
servers at night
 Greater application performance by allowing workloads to be moved
to less loaded servers
Workload Partitions Manager
 Increased administrative efficiency by federating management of
workload partitions across the IT landscape
 Improved IT efficiency though automated relocation of WPARs
 Greater availability through built-in monitoring of WPARs
 Also available as a component of AIX Enterprise Edition
10
© 2009 IBM Corporation
IBM Power Systems
New Manageability Capabilities
New capability
How capability addresses client needs
IBM Systems Director Console for
AIX
IBM Systems Director
AIX Enterprise Edition
AIX Management Edition
 Improved manageability by allowing administrators manage
AIX via a secure web interface
Integrates with Director 6.1 to provide seamless access
between Director and AIX contexts
Automated Page Size Management
for POWER6
 Improved performance management through AIX self tuning
page size on POWER6 systems
Workload Partitions
Live Application Mobility
Live Partition Mobility
Workload Partitions Manager
Role Based Access Control
AIX Security Expert
Trusted Execution
Concurrent AIX Updates
probevue dynamic tracing
 All of these features provide improved manageability by
reducing the effort of managing AIX, improving administrator
efficiency and reducing problem determination and resolution
11
© 2009 IBM Corporation
IBM Power Systems
Security
12
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 Role Based Access Control (RBAC)
What is it?
How it can help?
A new capability of AIX V6.1
that allows privileged
administration tasks to be
delegated to non-privileged
users
AIX
Resources
Users
Roles
DBA
Access to system resources
are associated with roles that
are assigned to non-privileged
users
Many roles are predefined
which can reduce the effort of
implementing RBAC
Roles can also be associated
with programs
BACKUP
Enables a more secure IT
infrastructure by reducing the
need for so many privileged
administrators
Create “create boot image”
Halt
“halt the system”
Info
“display boot information
Reboot “reboot the system”
Shutdown “shutdown the system”
Assigning roles to programs can
reduce the need for security
exposures such as the use of
setuid for programs
PRINT
aix
device
fs
network
proc
ras
security
system
wpar
boot
config
install
stat
auth = aix.system.boot.create
13
Can reduce the cost and
complexity of security
administration by allowing secure
delegation of administrative tasks
to non-privileged users
Allows for new ways to delegate
administration duties between
system administrators and nonadministrative users
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 Encrypting Filesystem
What is it?
The capability to automatically
encrypt data in a JFS2
filesystem
Data can be protected from
access by privileged users
How it can help?
Login Authentication Module
CLiC
User and Group
Key Stores
Crypto Lib
Automated key management key store open on login,
integrated into AIX security
authentication
Each file encrypted with a
unique key
No keys stored in clear in
kernel memory
Backup/Restore
Mgt Cmds
Cp, mv, crfs, etc
Data in clear in memory.
VMM
Crypto Kernext
Backup in encrypted or clear
formats
BOS Cmds
Key Store
Kernel ucred open
key store
J2
Filesystem
Always encrypted on disk
Enables improved security by
reducing unauthorized access to
data, even by privileged users
Secure backups reduces the
exposure of data compromised
when backup media is taken
outside of secure facilities
Automatic management of
protection keys can reduce the
administrative effort of using
encrypted data
Provides the capability for
additional security for applications
that may have security design
exposures
A variety of AES, and RSA
cryptography keys supported
14
© 2009 IBM Corporation
IBM Power Systems
AIX & Power Systems Security Certifications
AIX 5200-06 CAPP/EAL4+
Application: 01/11/05
Final report: 10/26/05
Certificate: 12/14/05
AIX 5L 5200-05 and
Pitbull LSPP/EAL4+
Application :01/11/05
Certificate issued: 05/16/06
AIX 6100-00)
CAPP/RBACPP/LSPP/EAL4+
MLS capabilities integrated into
standard AIX product
One certification for 3 Protection
Profiles
Supports P6, P5, P4
Certificate issued: May 2008
AIX 5300-05
LSPP/EAL4+
Pitbull MLS Ported to
AIX 5300-03
Pitbull product Supports P5, P4
Certificate issued: 12/19/06
Pitbull product available to
customers Dec 31, 05
Certification History
AIX 4.2 C2: Apr 24, 1997
AIX 4.3 C2: May 6, 19987
AIX 5.2 CAPP/EAL4+ : Nov 4, 2002
POWER4 HW CAPP/EAL4+ : Apr 2003
AIX 5.2 ML1 CAPP/EAL4+ : Sept 8, 2003
AIX 5.2 ML6 CAPP/EAL4+ : Dec 14, 2005
AIX 5.2 ML5 and Pitbull LSPP: May 16, 2006
AIX 5.3 TL5 and Pitbull LSPP: May 16, 2006
AIX 5.2 TL4 & VIOS CAPP/EAL4+: Dec 16, 2006
POWER6: Dec, 2007
AIX 6: May, 2008
15
AIX 5300-04 CAPP/EAL4+
Supports P5, P4
Certificate issued: 12/19/06
VIOS EAL4+
Included with AIX 53.00-04
CAPP/EAL4+
POWER6 Hardware
EAL4+
Dynamic LPAR with
MicroPartitioning
Legend
AIX V5.2
AIX V5.3
AIX V6.1
VIOS
POWER6
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 Security Expert
What is it?
A centralized security
management tool that can
control over 300 security
settings from a single console
Administrators can start from a
“Low”, “Medium”, “High” or
“Sarbanes-Oxley” security
template and customize settings
to met business requirements
Security settings can be
exported and imported as a
security profile to multiple
systems
On AIX V6.1, security profiles
can be stored in an LDAP
directory for ease of distribution
AIX Security Expert was first
included in AIX V5.3 TL5
16
How it can help?
Can reduce the cost and
complexity of security
administration by allowing
federated management of security
profiles across multiple servers
Enables a more secure IT
infrastructure by reducing the
effort of maintaining system
security
“Check” functionality can provide
additional security by validating
that the security profile for each
system matches the actual
security settings
Allows for new ways to
efficiently manage
security across
multiple AIX systems
© 2009 IBM Corporation
IBM Power Systems
Compliance Challenges
Companies face increased pressure to achieve and maintain
compliance – all with limited resources, time and budget.
Analysts have estimated that North American companies are
spending:
$29.9B on regulatory compliance
 $8.8B on technology solutions
Businesses are looking for compliance automation solutions
that provide:
Configuration in large scale enterprises
Single solution for consolidating and automating multiple
compliance regulations and standards.
Audit reporting to satisfy disparate compliance
organizations
Technology solutions provide automation to enable efficiency and improve
IT governance
 Payment Card Industry Data security standards and profiles
 DOD Security implementation Profiles

43% of CFOs think
that improving
governance, controls
and risk management
is their top challenge.
64% of CIOs feel that
the most significant
challenges facing IT
organizations are
security, compliance
and data protection
CFO Survey: Current state & future direction,
IBM Business Consulting Services
IBM Service Management Market Needs
Study, March 2006
17
17
© 2009 IBM Corporation
IBM Power Systems
IBM Compliance Expert Express Edition
IBM Compliance Expert Express Edition is designed to simplify IT compliance
Features:
Easily set dozens of AIX security configuration settings to match compliance standards
Includes profiles with recommended system settings for:
The Payment Card Industry Data Security Standard Version 2
The US Department of Defense Security Technical Implementation Guide for UNIX
Simple command line interface minimizes training and administrative workload
Reports that show whether the system configuration matches the compliance standard
Support for AIX 6 and AIX V5.3 on current Technology Levels
Potential Benefits:
Designed to simplify the effort of maintaining system configuration for compliance
Preconfigured profiles facilitate standardization and easy implementation and potentially reduce
the amount of administrative effort to interpret standards
Compliance reports may be used to provide a basis for audit activity
Note: Almost all compliance standards include procedural elements that are outside the
scope of system configuration settings. The IBM Compliance Expert Express Edition can
potentially simplify compliance efforts, but it cannot, by itself, enforce compliance.
18
© 2009 IBM Corporation
IBM Power Systems
IBM Compliance Expert Express Edition
Simplifies IT compliance with industry security standards
New
Automatically sets many AIX security settings to match common compliance
standards
Includes profiles with recommended system settings for:
The Payment Card Industry Data Security Standard Version 2
The US Department of Defense Security Technical Implementation Guide for
UNIX
Simple command line interface minimizes training and administrative workload
Reports show whether the system configuration matches the compliance standard
Support for AIX 6 and AIX V5.3 on current Technology Levels
Lowers cost of system administration for compliance standards
Facilitates standardization with minimal training
Manages risk with easy-to-use reports as a base for compliance audits
19
© 2009 IBM Corporation
IBM Power Systems
Other Security Features
 Secure by Default - Secure by Default takes a bottom-up approach in
hardening an AIX system by installing a minimal set of software, This
approach is opposite to starting with a regular, full-blown AIX installation and
then use the AIX Security Expert to apply hardening (top-down approach) by
disabling unneeded components.
 Trusted AIX - Enables Multi Level Security (MLS) capabilities in AIX. As
compared to regular AIX, Trusted AIX label-based security implements
labels for all subjects and objects in the system.
 The Trusted Execution environment - Enhances the AIX security
environment. It is a collection of features used to verify the integrity of the
system and implement advance security policies, which together can be
used to enhance the trust level of the complete system with Trusted
Signature Database (TSD).
 Trusted Execution - Provides a new command to verify the integrity of
the system. The trustchk command used for integrity checking: System
integrity check and Runtime integrity check.
20
© 2009 IBM Corporation
IBM Power Systems
Other Security Features
 Password length and encryption algorithms - Loadable
Password Algorithm (LPA). It also removes the eight
character password limitation.
 Discretionary access control - DAC are the security
aspects that are under the control of the file or directory
owner.
 FPM – File Permission Manager manages the permissions
on commands and daemons owned by privileged users with
setuid or setgid permissions.
21
© 2009 IBM Corporation
IBM Power Systems
Reliability
Availability
Serviceability
22
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 POWER6 Storage Keys
Before POWER6 Storage Keys
UNIX Kernel Address Space
 Exploitation of a POWER6
processor hardware feature to
provide additional isolation of
kernel and application data
 Storage keys can prevent invalid
changes to memory cause by
programming errors
 Application use of POWER6
storage keys is enabled in AIX
V5.3
 AIX kernel exploitation of
POWER6 storage keys is included
in AIX V6.1
23
AIX Kernel
JFS2
LVM
VMM . . .
Application
Address Space
AIX Drivers
SCSI
ENT
FC
Kernel
Code
WS
DB2
User
Code
User
Data
Kernel
Data
Files
After POWER6 Storage Keys
AIX Kernel Address Space
AIX Kernel
JFS2
LVM
VMM . . .
Application
Address Space
AIX Drivers
SCSI
ENT
Kernel
Code
Kernel
Data
FC
WS
DB2
User
Code
User
Data
Files
Can provide for higher AIX
availability by reducing the
number of unplanned outages
due to intermittent memory
overlay
All statements regarding IBM's future direction and intent are subject
to change or withdrawal without notice, and represent goals and objectives only.
© 2009 IBM Corporation
IBM Power Systems
AIX 6.1 Concurrent Maintenance
vmmove()
vmmove()
sleepx()
getgidx()
Kernel Space
User Space
emgr
Concurrent
update
vmmove() patch
Interim Fix
Fix selected AIX kernel problems without a service outage
Non-disruptive fixes to executable code in a running AIX kernel
 Base AIX Kernel (/unix), kernel extension, or device driver
No downtime (reboot) required to apply fix and make it active
Concurrent updates will be packaged as Interim Fixes
Maintenance can be backed off without an outage
24
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 POWER6 Automatic Variable Page Size
What is it?
AIX V6.1 exploitation of a
POWER6 feature that supports
variable page size
AIX will automatically select
optimal page size to provide
better performance
Kernel will choose between 4K
and 64K pages, including a mix
within a memory region
Supports process data, heap,
stack, shared memory,
anonymous mmap() memory
Enabled by default with
administrative controls to turn
off or change aggressiveness to
“upsize”
25
How it can help?
Can improve overall system
performance which could improve
the amount of work done per Watt
of energy
Automated page size tuning can
reduce the amount of effort and
cost associated with managing a
key aspect of performance tuning
Since this feature is turned “on”
by default, it improves your ability
to get the most out of your
systems based on POWER6
processors
This “self tuning” aspect of AIX
V6.1 can improve performance
while reducing administrative
workload
© 2009 IBM Corporation
IBM Power Systems
Cluster Aware Performance Utilities
 Benefits
– Group resource utilization
– Monitor multiple partitions in a single screen
Features
– AIX topas utility is made cluster aware. Cluster
aggregated utilization along with individual
partition statistics are displayed
– Recognizes PowerHA clusters
automatically
– User-defined group of hosts can be
monitored as a cluster
– Similar look & feel of CEC Monitor
Panel
– AIX topasrec utility is made cluster aware
– Records Utilization aggregated at
Cluster Level along with individual
partitions utilization
– AIX topasout utility is used to process the
cluster recording file & generate reports
– Reports that can be post processed by
nmon analyzer
26
© 2009 IBM Corporation
IBM Power Systems
AIX Runtime Expert:
Providing simplified discovery, application, update, and verification of O/S
runtime properties across one or more systems
Old Way
New Way
Environment Variables
Boot LV Settings
OS Configuration and Tuning Soup
Control 1
Config file x contents
Env var XYZ=“Yes”
AIX security profile T
Single
XML Profile
.
tuneable N
CLI Utilities
Configuration Files
Set
Apply and maintain approaches
Scripts, ftp, rsh, ssh, documentation,
3rd party tools, mksysb, etc.
“One
Button”
Tasks
Extract
Act on
Multiple
Systems
Compare
System A
27
System B
System N
System
© 2009 IBM Corporation
IBM Power Systems
AIX Runtime Expert
(included with AIX 6.1 TL4)
New
Enables the administrator to set, most
AIX tuning and configuration
settings with confirmation profile
Existing systems can be used as a
model to create the profile
Current system settings can be
compared to a profile to detect
unauthorized changes
Extensible architecture with a
programmable core engine to
enable system administrator
customization
Control 1
Config file x contents
Env var XYZ=“Yes”
AIX security profile T
Single
XML Profile
.
tuneable N
Set
“One
Button”
Tasks
Extract
Compare
Planned to be integrated with
Systems Director in 2010
28
© 2009 IBM Corporation
IBM Power Systems
AIX Runtime Expert - Architecture
Extensible architecture with a programmable core engine
Base Command Line Utilities
artexget – extract runtime attributes from a running system based on a provided configuration profile
artexset – set values on a system from a profile to take effect immediately or after system restart
artexdiff – compare values between a running system and a profile, or compare between two profiles
artexmerge – combine the contents of two or more profiles into a single profile
artexlist – list configuration profiles that exist on a system
Set
Get
Configuration Profiles
•XML files
Core Engine
•Based on Control Catalogs
•Supplies values
Running O/S
Instance
Diff
Merge
•Template to get (extract) values
•Can reside in LDAP or locally
Control Catalogs
•XML files
•Programming Modules for the Core Engine
•Specifies parameter-value rules, processing
sequences, environment variable details,
file content management, etc.
29
© 2009 IBM Corporation
IBM Power Systems
AIX Runtime Expert – Configuration Elements
acctctl
dumpctrl
nis
namerslv
tsd
alog
errdaemon
probevue
nfs
trustchk
authzcfg
ewlm
tcp_nw
shconf
vmo
authent
ffdc
udp_nw
schedo
aix.secexpert
chcons
filter
ip_nw
privcmd
mkuser.defuser
Chdev.sys0
ioo
arp_nw
privdev
chuser
chlicense
krecovery
stream
privfile
login
chservices
lvmo
raso
smtctl
chsubserver
chsys
nfso
role
syscorepath
gen.param
class
mktcpip
ruser
traces
etc.env
sysdumpdev
file.data
trcctl
restricted
misc.other
probeview
30
© 2009 IBM Corporation
IBM Power Systems
Memory Pagesize tuning
 Options in vmo command
 vmm_default_pspa
-1: no page support or no hardware support
 0-100: Percent of pages unreferenced before
promotion occurs
Defaults:
• -1: for Power5 and before servers
• 0: for Power5+ and above servers
 vmm_mpsize_support
0: AIX recognizes only 4K & 16MB pages
1: AIX use pagesizes supported by processor
2 (default): AIX use multiple page size per segment
 pagesize –af: Display all supported virtual memory page sizes
31
© 2009 IBM Corporation
IBM Power Systems
Performance tuning Changes
KERNEL/PROCESS/TUNING
5.1
5.2
5.3
6.1
32,767
32,767
32,767
32,768
Limit max Threads/process and max processes/user
(chuser and ulimit commands)
N
N
N
Y
AIXTHREAD_SCOPE – Note 45
P
P
P
S
8:1
8:1
8:1
1:1
ksh93/ ksh88
ksh93/ ksh88
ksh93/ ksh88
ksh93/ ksh88
5K
25K
25K
Max Threads/process
AIXTHREAD_MNRATIO– Note 45
ksh
Max. no. of devices – Note 32
Memory/Storage Keys Application
N
N
Y-P6
Y-P6
Memory/Storage Keys Kernel
N
N
N
Y-P6
Restricted tuneables – Note 34
N
N
N
Y
Out of the box Perf Tuning VMM, AIO and Oracle – Note36
N
N
N
Y
Solution Performance Tuning – Note 36
N
N
N
Y
System
System
Filesystem
Filesystem
I/O pacing enabled by default
N
N
N
Y
aio_minservers
1
1
1
3/core
aio_maxservers
10
10
10/core
30/core
aio_maxrequests
4096
4096
4096
65536
aio_fastpath & aio_fsfastpath
0/0
0/0
0/0
1/1
minpout/maxpout
0/0
0/0
0/0
4096/8192
20/80/80
20/80/80
20/80/80
03/90/90
lru_file_repage
page_steal_method
1
0
1
0
1
0
0
1
Memory Affinity (To disable vmo -o memory_affinity=0)
N
Y
Y
Y
nmon integration into topas
Y
Y
VIOS
Y
Y
I/O Pacing tuning level
minperm/maxperm/maxclient
REFERENCE: Section 6.3 / AIX 6.1 Differences
Redbook,
N
N SG24-7559
http://www.redbooks.ibm.com/abstracts/sg247559.html?Open
monitoring
in topas
N
N
32
© 2009 IBM Corporation
IBM Power Systems
Increased Argument Size
 AIX 5.3 was capped at 24KB
 AIX 6.1 supports 24KB to 4MB
 ARG_MAX increased from 256K 1024K in limits.h
 To change
view with “lsattr –El sys0 –a ncargs”
change with “chdev –El sys0 –a ncargs=n
 where n is number of 4K blocks
 chdev –El sys0 –a ncargs=256 # results in 1MB size
(256 * 4K)
REFERENCE: Section 5.3 / AIX 6.1 Differences Redbook, SG24-7559
http://www.redbooks.ibm.com/abstracts/sg247559.html?Open
33
© 2009 IBM Corporation
IBM Power Systems
Network Changes
 IGMPv3 implementation adheres to RFC 3376 and includes the new Socket Interface
Extensions for Multicast Source Filters.
 IGMPv3 implementation allows backward compatibility with the previous two versions of
the protocol, IGMP version 1 (IGMPv1) and IGMP version 2 (IGMPv2)
 IGMPv3 protocol supports two distinct multicast modes: Any-source multicast (exclude)
and Source-specific multicast (include) mode.
 NDAF (Network Data Administration Facility) enhancements
 A secure version of ftp (and ftpd), based on OpenSSL, using Transport Layer Security
1(TLS) to encrypt both the command and the data channel.
 NFS Proxy serving enhancements
 NFS backwards support – NFSv3 exports for back-end NFSv4 exports
 New network caching daemon (netcd) to improve performance for resolver lookups.
Netcd can cache user and group information provided by a NIS server.
 IPv6 RFC compliances - compliant with RFC 4007 and RFC 4443
34
© 2009 IBM Corporation
IBM Power Systems
Others - Support for USB Drives
 AIX 6.1 TL2 supports USB Drives (Flash/Thumb Drives)
Requires filesets
 devices.common.IBM.usb
 devices.usbif.08025002
 Configuration
“cfgmgr –l usb0” gets two devices: /dev/flashdrive0 &
rflashdrive0
Use archive commands - tar, backup/restore, cpio or dd
Use ISO filesystem
• “mount -rv cdrfs /dev/flashdrive0 /mnt”
• ISO file system (created by mkisofs) is written to flashdrive with
dd command: “dd if=myimage.iso of=/dev/flashdrive0”
35
© 2009 IBM Corporation
IBM Power Systems
Virtualization
Features
36
© 2009 IBM Corporation
IBM Power Systems
PowerVM Virtualization Architecture
Management and Provisioning
AIX V6
partitions
AIX / WPARS
Kernels
Hardware
Management
Console*
(HMC)
Hypervisor
AIX V5.3
partitions
Linux
AIX
Kernels
Linux
Kernels
IBM i
Virtual
I/O
Server
SLIC
VEnet
VSCSI
IVM**
Virtual /Networks
Virtual Network
Storage
Virtual Processors
Unassigned
on demand
resources
Virtual Storage
Processors
Service
Processor
Memory
Expansion slots
Local devices and storage
Networks and network storage
*Integrated Virtualization Manager (IVM) is disabled if HMC attached
37
© 2009 IBM Corporation
IBM Power Systems
Power Systems Virtualization with PowerVM
Dynamically Resizable
Virtual
1
I/O
Cores
Server
Partition
2
Cores
Int Virt
Linux
Linux
3
3
6
Cores
CoresCores
Storage
Sharing
Ethernet
Sharing
Virtual I/O paths
AIX
V6.1
Linux
AIX
V5.3
IBM i
AIX V5.3
AIX V6.1
AIX i
IBM
V6.1
AIX V5.3
Micro-partitioning
AIX V6.1
Linux
Linux
Manager
5
2
Cores Cores
Micro-Partitioning Feature
Share processors across
multiple partitions
Minimum partition 1/10th core
254 partition maximum
AIX V5.3/6.1, Linux, & IBM i
Managed via HMC or IVM
Virtual I/O server
Virtual LAN
POWER Hypervisor
Network
Shared Ethernet
Shared SCSI & Fibre Channel
attached disk subsystems
IVM
Web
Browser
38
© 2009 IBM Corporation
IBM Power Systems
AIX 6 Workload Partitions (WPARs)
S/W Partitioned system capacity
Each WPAR gets a regulated share of the processor
and memory resources
Each WPAR has separate network and filesystems
and many system services
Separate Administrative control
Workload
Partition
dB
Each partition is a separate
administrative and security domain
Shared system resources
I/O Devices
Processor & Memory
Operating system
Shared Library and Text
Workload
Partition
App
Workload
Partition
Web
WPAR Types: System and Application
Workload
Partition
ERP
Workload
Partition
Test
System has own copy of /, /tmp, /var, /home
Application uses Global filesystems
Benefits
AIX 6
Separate regions of application space within a single
AIX image
 Improved administrative efficiency
 Reduced no. of OS images to administer and
maintain
System commands are WPAR enabled
39
© 2009 IBM Corporation
IBM Power Systems
WPAR Shared Applications Enables Administrative
Efficiency
Application installed in Global instance and used by multiple WPARs
Global FS
Global filesystems
/
/etc
/usr
application code
/opt
application code (or here)
/var
/tmp
/appserver application code (or here)
System WPAR filesystems
/
r/w - unique per WPAR
/etc
r/w - unique per WPAR
/usr
r/o from global (typically)
/opt
r/o from global (typically)
/var
r/w - unique per WPAR
/tmp
r/w - unique per WPAR
/appserver r/o from global
/config
r/w uniq per WPAR (example)
AIX
/
/etc
/usr
/opt
/var
/tmp
/appsvr
WPAR FS
App Server 1
/
/etc
/var
/tmp
/config
WPAR FS
App Server 1
global Instance
Workload
Partition
Workload
Partition
App Server
#1
Billing
Workload
Partition
App Server
#2
Workload
Partition
Workload
Partition
BI
Web
Server
/
/etc
/var
/tmp
/config
NFS
40
© 2009 IBM Corporation
IBM Power Systems
AIX Workload Partitions Can Be Used in LPARs
Dedicated
Processor
LPAR
Finance
Dedicated
Processor
LPAR
Planning
LPAR
LPAR
Americas
WPAR #1
MFG
WPAR #1
Business
Intelligence
VIO
Server
LPAR
Asia
LPAR
EMEA
WPAR #1
eMail
WPAR #2
Test
WPAR #2
Planning
WPAR #3
Billing
Micro-partition Processor Pool
POWER Hypervisor™
41
© 2009 IBM Corporation
IBM Power Systems
Flexible PowerVM & AIX Options




Dedicated LPARs
Micro-Partitions
WPARs
Workload Manager
- Application Mobility
- Partition Mobility
LPAR /
Micropartitions
Resource Flexibility &
Ease of Administration
AIX V5.3 on POWER5 or later
Workload
Partitions
Live Partition
Mobility
AIX 6 on POWER4 or later
AIX Workload
Manager
Live Application
Mobility
AIXV4.3.36 on POWER3 or later
Workload Isolation
42
© 2009 IBM Corporation
IBM Power Systems
Live Application Mobility….
Dedicated Dedicated
Processor Processor
LPAR
LPAR
Planning
Finance
Micro-Partition Processor
Pool
Micro-Partition
Processor
Pool
LPAR
LPAR 1
LPAR 2
LPAR 3
Micro-Partition
Processor Pool
LPAR 1
LPAR 2
LPAR 3
VIO
Server
Power Hypervisor
Power Hypervisor
Move live Workload Partitions between physical systems ( Common Hardware )
Workloads move, not the whole partition
Partition OS images must be the same ( Service Level )……
43
© 2009 IBM Corporation
IBM Power Systems
AIX 6 Live Application Mobility
NFS
AIX # 1
Workload
Partition
ERP
Workload
Partition
Database
Workload
Partition
Web
AIX # 2
Workload
Partition
QA
Application
Partition
Dev
Workload
Partition
Billing
Workload
Partition
Manager
Workload
Partition
Data Mining
Policy
Move a running Workload Partition from one server to another for outage avoidance
and multi-system workload balancing
 Works on any hardware supported by AIX 6, including POWER5 and POWER4
44
© 2009 IBM Corporation
IBM Power Systems
Workload Partitions Manager
 Management of WPARS across multiple systems
 Lifecycle operations
 Single Console for:
Graphical Interface
Create & Remove
Start & stop
Checkpoint & Restart
Monitoring & Reporting
Manual Relocation
Automated Relocation
Policy driven change
Workload
Partition
Manager
 Infrastructure Optimization
 Load Balancing
Web Service
Global Level
Global Level
Global Level
WPAR Agent
WPAR Agent
WPAR Agent
System/Application WPARs
45
Browser
System/Application WPARs
System/Application WPARs
© 2009 IBM Corporation
IBM Power Systems
PowerVM Live Partition Mobility with POWER6
Allows migration of a running LPAR to another physical server
with no application downtime
 Reduce impact of planned outages
 Relocate workloads to enable growth
 Provision new technology with no disruption to service
Movement to
a different
server with
no loss of
service
Virtualized SAN and Network Infrastructure
46
© 2009 IBM Corporation
IBM Power Systems
Partition Mobility: Active and Inactive LPARs
Active Partition Mobility
 Active Partition Migration is the actual movement of a running LPAR from one
physical machine to another without disrupting* the operation of the OS and
applications running in that LPAR.
 Applicability
 Workload consolidation (e.g. many to one)
 Workload balancing (e.g. move to larger system)
 Planned CEC outages for maintenance/upgrades
 Impending CEC outages (e.g. hardware warning received)
Inactive Partition Mobility
 Inactive Partition Migration transfers a partition that is logically ‘powered off’
(not running) from one system to another.
Partition Mobility supported on POWER6™
AIX 5.3, AIX 6.1 and Linux
47
© 2009 IBM Corporation
IBM Power Systems
Mobility on Power Systems
Live Partition Mobility
Movement of the
OS and
applications to a
different server
with no loss of
service
Virtualized SAN and Network Infrastructure
PowerVM Live Partition Mobility
• Move an entire Logical Partition from one system to another
while it is running with almost no impact to end users
• Moves the entire LPAR including the operating system
• Requires systems based on the Power6 processor, PowerVM
Enterprise, and all I/O must be through the Virtual I/O Server
• Works with partitions running AIX V5.3, AIX 6 and Linux
Potential Benefits
Improved application availability
Energy saving
Better workload management
Live Application Mobility
48
AIX Live Application Mobility
• Move a Workload Partition from one AIX system to another
AIX system while running with almost no impact to end users
• Moves only the WPAR, the AIX operating system is not
moved
• Requires AIX 6, PowerVM Workload Partitions Manager, and
all WPAR filesystems must be NFS
• Works on systems based on Power4, Power5, and Power6
processors
© 2009 IBM Corporation
IBM Power Systems
Active Memory Sharing
Pools of memory that can be shared by partitions
Similar to shared processor partitions
 Pool of processor resources
OS Support
AIX 6.1, IBM i and Linux
Features
Allows for the dynamic sharing of memory
Provides the ability to “Over-Commit” physical memory
 Overflow of memory request paged to system disk.
Fine-grained sharing of physical memory
Automated ballooning (expansion and contraction) of a partition’s physical
memory footprint based on workload demands.
Sharing of common code pages between partitions
 Reduces the memory and cache footprints
 Partitions with the same OS and application code.
49
© 2009 IBM Corporation
IBM Power Systems
Active Memory Sharing Overview
Active Memory Sharing intelligently flows memory from one partition
to another for increased utilization and flexibility of memory usage
Memory virtualization enhancement for Power Systems
Memory dynamically allocated based on partition’s workload demands
Contents of memory written to a paging device
Improves memory utilization
Designed for partitions with “Variable Memory” requirements
Low average memory requirements
Active / Inactive environments
Workloads that peak at different times across partitions
Available with PowerVM Enterprise Edition
AIX 6.1, Linux, and IBM i 6.1 partitions that use VIOS and shared processors
POWER6 processor-based systems
All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
50
© 2009 IBM Corporation
IBM Power Systems
Dedicated vs Active Shared Memory Environment
Partition 1
30
Partition 2
Partition 3
Partition 1
Partition 4
Dedicated Memory
30
25
25
20
20
15
15
10
10
5
5
0
0
Time
51
Partition 2
Partition 3
Partition 4
Partition 5
Active Shared Memory
Time
© 2009 IBM Corporation
IBM Power Systems
Active Memory Sharing Requirements
Hardware assist available in POWER6, POWER7, etc.
Redirects Data Storage Interrupts (DSI) & Instruction Storage Interrupts
(ISI) to the Hypervisor
Allows the Hypervisor to relocate a partition’s physical memory pages with
finer granularity than a Logical Memory Block (LMB)
Active Memory Sharing is enabled on a partition by partition basis
 Partitions must be defined in Shared Processor Pool
 Partitions must have all virtual resources, no real I/O
No support for pages >16MB
Supported Operations Systems:
 IBM i V6R1 PTF
 AIX 6.1 TL3
 Linux SLES 11 and RHEL 6.0
PowerVM Enterprise Offering
VIOS 2.1.1
HMC Firmware 7.342
FW release eFW3.4.2
Disruptive update from previous FW release
52
© 2009 IBM Corporation
IBM Power Systems
System
Management
53
© 2009 IBM Corporation
IBM Power Systems
AIX V6.1 Systems Director Console for AIX
What is it?
A new Web based
management tool that provides
easy access to common system
administration tasks
Administrators can access
Systems Management Interface
Tool (SMIT) menus from a
browser
Graphical user interface is fast
and fully integrated with IBM
Systems Director
All necessary components for
the Console are included in AIX
The Distributed Command
Execution Manager (DCEM)
feature of the Console allows an
administrative task to run on
multiple systems at once
54
How it can help?
Can reduce the amount of effort
and cost associated with
managing the AIX OS
Web access to administrative
tasks can simplify systems
management
Consistent user interface with
IBM Systems Director and the
WPAR Manager can reduce
retraining and other administrative
costs
The combination of Web access
to administration tools and the
ability to execute administrative
tasks on multiple systems can
change the way you manage the
AIX OS
© 2009 IBM Corporation
IBM Power Systems
Systems Director Console for AIX (pconsole)
 Systems management via a Browser
 It is SMIT in a Browser
 Works with IE 6.0+ or Firefox
 Installed and configured by default
 sysmgt.pconsole.rte filesets
 To launch use the URL below
 https://ipaddress:5336/ibm/console/login.jsp or
https://ipaddress/ibm/console
REFERENCE: Section 5.7 / AIX 6.1 Differences Redbook, SG24-7559
http://www.redbooks.ibm.com/abstracts/sg247559.html?Open
55
© 2009 IBM Corporation
IBM Power Systems
Smooth Upgrade to AIX 6
 No charge upgrade for current AIX 5L clients with SWMA
 No additional out of pocket expense for clients
 Upgrade process
 Tools like alt disk installation and NIM minimize client risk
 Migration installation from AIX V4 & AIX V5 supported
 New GUI installation tool, when AIX Installation DVD media
is used for booting
56
© 2009 IBM Corporation
IBM Power Systems
AIX Editions
57
© 2009 IBM Corporation
IBM Power Systems
AIX Editions…..
AIX 5.3 Management Edition bundle
AIX V5.3
Tivoli® Application Dependency Discovery Manager
IBM Tivoli Monitoring
IBM Usage & Accounting Mgr Virtualization Edition for Power Systems
AIX 6.1 Enterprise Edition
AIX V6.1
PowerVM AIX Workload Partitions Manager
Tivoli® Application Dependency Discovery Manager
IBM Tivoli Monitoring
Tivoli Performance Analyzer
58
© 2009 IBM Corporation
IBM Power Systems
AIX Enterprise Edition Key Features
Live Application Mobility
Relocate Workload Partitions between systems with almost no client impact
Manage WPARs across multiple systems
Centralize the creation, replication, and starting of WPARs across multiple systems
Automatically discover IT components and their relationships
Ideal for managing dynamic virtualized environments
Monitor virtualized resources
Efficient management begins with comprehensive performance information
Predictive monitoring and capacity management
Leverage real time monitoring in data warehouse to provide advanced analytics for capacity planning and
proactive monitoring
Provides a visual representation of the components
Assists understanding of complex application dependencies
Monitor utilization and configuration changes
Useful for problem determination and failure analysis
Tivoli Performance Analyzer
Extends ITM data to predictive trending to manage performance over time
Forecast resource trends to focus monitoring on emerging problems.
Leverages the long-term historical and real time data in Tivoli Data Warehouse
Collect and report resource usage
Understand IT resource consumption by workload or area
59
© 2009 IBM Corporation
IBM Power Systems
Simplified WPAR Management
60
© 2009 IBM Corporation
IBM Power Systems
AIX Editions Solution Components: Discovery
IBM Tivoli Application Dependency Discovery Manager (TADDM)
IBM Tivoli Application Dependency
Discovery Manager initiates and assists
planning for consolidation by providing
best-of-breed discovery capabilities
Discovers the COMPONENTS in a Data
Center Environment
CENTRALIZES and VISUALIZES the
CONFIGURATION of the Components in a
Data Center Environment
Discovers the RELATIONSHIP of the
Components in a Data Center Environment
DISCOVERS AND TRACKS THE
CHANGES in a Data Center Environment
Can Feed this Information to other IBM Tivoli® Products
61
© 2009 IBM Corporation
IBM Power Systems
AIX Editions– IBM Tivoli Monitoring
IBM Tivoli Monitoring helps prioritize consolidation decisions by
visualizing the actual virtual server utilization against historical
trends. It automates a clients best practices in response to system
events
Improves mean-time-to-recovery by
visualizing the virtual world to solve
“virtual performance problems”
Side-by-side real-time and historical data
assists in separating intermittent problems
from reoccurring problems from peak
workloads
Out-of-the-box reporting allows
clients to quickly provide executive
level reports and identify resource
bottlenecks
IBM Tivoli Monitoring
65
© 2009 IBM Corporation
IBM Power Systems
Power System Monitoring – Hypervisor View
Global
CPU &
Memory
allocation
Total CPU
& Memory
allocated
to LPARs
66
© 2009 IBM Corporation
IBM Power Systems
Power System Monitoring - AIX LPAR View
CPU, Memory,
Disk, Network
Info per LPAR
67
© 2009 IBM Corporation
IBM Power Systems
Monitoring - VIOS View Network / Disk Mapping / Utilization
Shows how
network
interfaces are
mapped to
LPARS
68
© 2009 IBM Corporation
IBM Power Systems
Overall Frame Utilization
69
© 2009 IBM Corporation
IBM Power Systems
CEC Event links to IBM Director
70
© 2009 IBM Corporation
IBM Power Systems
Advanced Performance Analytics
What It Does
• Provide capacity monitoring through the data collected by Tivoli Monitoring
• Automates Performance analysis and reporting
• Enables prediction of application bottlenecks and creation of alerts for potential service
threats.
Use existing ITM agents and data that are stored in
the Tivoli Data Warehouse
Create new metrics based on combining existing
date
Predictive trending and forecast reports
Pre-configured reports
Extensible
Scenarios
“What will my resources look like
tomorrow, next week and next month?”
“What IT resources should I worry
about?”
“Will I have enough capacity to get me
through Monday?”
71
© 2009 IBM Corporation
IBM Power Systems
Predictive Trending
CPU
Predictive trending on key performance indicators
 Linear trending model
 Configurable
 Simple, open and predictable
New Tivoli Monitoring attributes for use in charts
and situations
 Trend strength, trend direction
 Time to threshold, value in 7 days, 30 days
and 90 days
Use trend information in situations
 “I predict I have 2 weeks before I hit 95%
Disk Utilization and I am 70% confident and
its getting worse”
Leverages Tivoli Enterprise Portal
 Overlays to represent Trends
 Icons in Tables
72
Threshold
Predicted
CPU Violation
Predicted trend
Actual Monitor Data
Time
© 2009 IBM Corporation
IBM Power Systems
Using Trends in Forecast Reports
Projected values across all resources
Sort lists to identify future overloaded or under utilized servers
73
© 2009 IBM Corporation
IBM Power Systems
Usage IBM Usage & Accounting Manager
AIX Editions Solution Component – Virtualization
Edition
Apportion usage by account, department or organization
Accountability and usage tracking ensures optimized usage by
each department
Easily forecast growth by department to justify year-to-year
budget
changes
IT Expenses by Account
Single hardware system metrics and reports
Data collectors
AIX, Linux® and AIX Advanced Accounting –
 Processor, server, LPAR, I/O, and VIO
OS File System – allocated and used
Usability – Power System tailored:
Administration Console
JobRunner GUI
Reporting
Business Intelligence Reporting Tool Reports
Reports will be provided, with aggregation by
userid within a given server
“Pre-Defined” Accounting Schema
Export to spreadsheet, comma delimited, and CSB
74
© 2009 IBM Corporation
IBM Power Systems
Power System Usage Trend Sample Reports
Example of Resource
Usage Trend report over a
period of time
Available for AIX, AIX Advance
Accounting, Linux
75
© 2009 IBM Corporation
IBM Power Systems
IBM Usage and Accounting Manager
Know what IT costs (in shared and virtualized environments)
Helps businesses to understand the true costs of
their IT
 Who is consuming which resources?
 What are the true costs of these resources?
 How should costs be allocated for ROI or
chargeback?
Enables businesses to make informed decisions
about IT options and acquisitions
Facilitates chargeback accounting to bill internal or
external customers for their actual resource use
Tracks and analyzes resource utilization across the
entire enterprise
 Servers, storage, networks, applications, etc.
76
© 2009 IBM Corporation
IBM Power Systems
AIX Management Offerings Summary
Offering
Description
OS Support
OS included
AIX Enterprise Edition
AIX 6 plus the WPAR Manager, Tivoli TADDM, Tivoli
Monitoring and Tivoli Usage & Accounting
Virtualization Edition
AIX 6
Yes
AIX Standard Edition
The AIX operating system
AIX 6, AIX V5.3,
AIX V5.2
Yes
Management Edition for AIX
Tivoli TADDM, Tivoli Monitoring and Tivoli Usage &
Accounting Virtualization Edition
AIX 6 and AIX
V5.3
No
IBM Systems Director
Cross system platform management
AIX, Linux, i,
zOS, Windows
No
Performance Management
for Power Systems
Remote performance reporting especially for smaller
clients with limited IT teams
AIX 6, AIX V5.3
and IBM i
No
Cross systems management for WPARs and
enablement for Live Application Mobility
AIX 6
No
(AIX)
(PM for Power Systems)
PowerVM Workload
Partitions Manager for AIX
(WPAR Manager)
77
© 2009 IBM Corporation
IBM Power Systems
Operating System
Release / Service
Strategy
78
© 2009 IBM Corporation
IBM Power Systems
New Service and Release Strategy
The principal changes planned are:
•Twenty four months of support for each Technology Level
•Service for entire period is provided by PTF, Interim Fix, and/or Service Pack
•New hardware within the same family will be supported on previous Technology Levels
for ease of migration.
Announcement in 2Q06 for AIX V5.3 Technology Level 6 and later releases
New hardware support for old TLs
is only possible within the same
processor and technology family
Exploitation of new HW may require
the latest TL or even a new release
79
Degree of hardware change
On prior TLs On the latest
plus latest
Technology
Service Packs
Level
Processor Speed Increase Only
( No AIX Code Changes)
Yes
Supported
New Processor in Compatibility Mode
( N0 AIX Code Changes)
Yes
Supported
New Processor in Family
( Recognize New processor )
New IO
( New Device Driver )
New Technology
( Significant / Pervasive )
Yes
Supported
Yes
Supported
No
Yes
© 2009 IBM Corporation
IBM Power Systems
Release Strategy Transition* AIX
2006
2007
Spring
Fall
V5.3 only
2008
Spring
Fall
2009
Spring
Fall
Spring
2010
Fall
Spring
The new release strategy goes into effect
starting with AIX 5.3 TL6 in 1H07*.
CSP
SP
Technology
Level 4
AIX V5.3 TL4 and TL5 will be supported under the previous strategy.
CSP
AIX V5.2 will only be supported under the previous strategy.
SP
Technology
Level 5
SP
SP
SP
HP
SP
SP
HP
SP
SP
SP
HP
Technology
Level 6
Legend:
SP
HP
SP
SP
SP
SP
SP
Service Pack – may include new HW
Service Pack – AIX fixes only
SP
Concluding Service Pack – Last Service Pack
CSP
SP
HP
SP
HP
Technology
Level 7
Interim Fix
SP
SP
Support via Interim Fix, PTF, or Service Pack
Support via CSP + Interim Fix
SP
HP
SP
HP
SP
SP
HP
SP
SP
SP
HP
Technology
Level 8
New Technology Level - New HW/SW support
and hardware exploitation)
80
© 2009 IBM Corporation
IBM Power Systems
Differences From Previous Release Strategy
2007
Spring
2008
Spring
Fall
2009
Spring
Fall
Fall
2010
Spring
2011
Spring
Fall
CSP
Previous
Release
Strategy
SP
Technology
Level 6
SP
New
Release
Strategy
SP
SP
HP
SP
SP
HP
SP
SP
SP
SP
HP
Technology
Level 6
Difference
Previous
New*
Length of Service for a TL
Extended service via …
Concluding Service Pack?
12 months
CSP + Interim Fix only
Yes, start of extended service
4-6 weeks
AIX V5.2 & AIX V5.3
24 months
PTF, Interim fix or Service Pack
No longer used
Fixes + new HW support within same
Family
8-12 weeks
AIX V5.3 and future releases
Policy started with . . .
AIX V5.2 TL8, AIX V5.3 TL4
AIX V5.3 TL6
Hardware support via…
Latest Technology Level only
Exploitation via latest TL. Some hardware
support available via prior TLs plus a SP
5.3.0.0040
5.3.7.0040
Service Packs include…
Service Packs ship every…
AIX releases supported…
Version Release Mod. Fix
81
Only fixes
© 2009 IBM Corporation
IBM Power Systems
New Hardware Supportability*
On prior TLs
plus latest
Service Pack
On the latest
Technology
Level
Processor Speed Increase Only
(No AIX Code Changes)
Yes
Supported
New Processor in Compatibility Mode
(No AIX Code Changes)
Yes
Supported
New Processor in Family
(Recognize New Processor)
Yes
Supported
New I/O
(New Device Drivers)
Yes
Supported
New Technology
(Significant/pervasive)
No
Yes
Degree of hardware change
Note: Exploitation of new hardware features will require moving up to the
latest TL or in some cases, moving up to the next AIX release
82
*All statements regarding IBM's future direction and intent are subject
to change or withdrawal without notice, and represent goals and objectives only.
© 2009 IBM Corporation
IBM Power Systems
IBM Systems
Director
83
© 2009 IBM Corporation
IBM Power Systems
IT organizations face multiple challenges getting the
value of virtualization
Q. What are the major hurdles you faced in implementing virtualized servers at your organization?
29.8%
None
34.1%
27.1%
Tools/Management
4.6%
20.7%
Vendor Support
7.8%
2007
19.8%
20.2%
In-house Expertise
2006
19.5%
21.9%
Technical Issues
17.8%
Time
32.3%
17.1%
Costs/Budget
27.4%
15.6%
Institutional Resistance
22.5%
7.8%
High Availability
0%
Source: IDC Virtualization Study, 2007
84
5%
10%
n=410
15%
20%
25%
30%
35%
40%
Percent of Sample
© 2009 IBM Corporation
IBM Power Systems
End-to-end Management Approach
IBM Director
Enterprise
Management
AIX Management
IBM Systems Director
Physical and virtual platforms
Server, Storage, Networking
Foundation
Health
Virtualization
Optimization
Configuration
Maintain
Advanced
Monitoring
Replication
Virtualization
software
Hardware
System
x, i, z, p
System
Storage™
3rd Party,
Custom
Other
And more . . .
Operating
systems
85
Extension Groups
Deployment
Platform-specific
capabilities
Managed
environments
Active Energy
Manager
© 2009 IBM Corporation
IBM Power Systems
IBM Systems Director for Power Topology
IBM Systems
Director Console
IBM Systems Director
Foundation
Management
Inventory
Health
Config
Update
Physical and virtual platforms
Server, Storage, Networking
AIX
Advanced
Management
Availability
Workload
Image
Energy Mgt
MM
WPAR
IBM i
WPAR
AIX
SLES RHEL VIOS
AIX
SLES RHEL IBM i
WPAR
JS21
HS21
SLES RHEL
AIX
Win
WPAR
HMC
PHYP
FSP
IVM
(VIOS)
POWER6 System
AIX (SMP)
PHYP
Blade Center
BMC/
FSP
PHYP
FSP
POWER 5 System
HMC: Hardware Management Console
WPAR: Workload Partition (Container)
IVM: Integrated Virtualization Manager
PHYP: POWER Hypervisor
VIOS: Virtual IO Server (virtual IO and Layer 2 bridge)
BMC: Baseboard Management Controller
MM: Management Module
SMP: Symmetric Multi Processor
FSP: Flexible Service Processor
86
© 2009 IBM Corporation
IBM Power Systems
Director 6.1 Power Capabilities
 Automatic Discovery and Inventory:
 POWER resources and connected storage which includes the
collection of both hardware and software inventory.
 HMC, CEC, LPAR, AIX, pLinux, HMC, VIOS, FSP, and Virtual
Networking components – bridges and VLANS)
 Visualize various POWER resource:
 Topologies and relationship across physical server and virtual servers
 CEC, HMC, VIOS, LPAR, devices, AIX, pLinux and Virtual
Networking components, virtual disks, logical volumes and
associated volume groups.
 Discovery and documentation of full system configuration
 Physical and virtual IO resources and association/relationship (for
configuration recovery – i.e. System plan).
87
© 2009 IBM Corporation
IBM Power Systems
Director 6.1 Power Capabilities
 Show Health and Status of Physical and Virtual Servers
 HMC and VIOS.
 Show Alerts: Hardware failures and system logs from VIOS, HMC
and the operating systems.
 Base Monitoring:
 OS Metrics: CPU and memory utilization
 File system metrics across hosts and virtual servers.
 Historical and OS events monitoring.
 View CPU utilization metrics for environments that contain both
shared and dedicated processors for both host and virtual servers.
88
© 2009 IBM Corporation
IBM Power Systems
Director 6.1 Power Capabilities
 Download, Manage, and apply recommended Updates
 AIX, pLinux, i5OS, HMC and System Firmware
 Deployment/Provisioning/Planning
 Ability to configure new systems or clone systems using system
plans Deployment of OS and VIOS on a LPAR via HMC.
 Base Virtualization Management
 Support key lifecycle LPAR and mobility operations
 Within single HMC domain) operations.
 Consolidated Interface
 Integration of tasks for Key Power Resource Managers
 HMC, IVM/VIOS, AIX and i5OS management consoles
89
© 2009 IBM Corporation
IBM Power Systems
Director 6.1 Power Capabilities
 Comprehensive CLI interface
 Discovery/Health/Update/Deployment, LPAR virtualization lifecycle
and mobility, power control and management
 Energy Management – Active Energy Manager
 Monitoring, reporting, capping (both a server and group), and
controlling power consumption.
 Receive power status and alerts.
 Energy Thresholding - Allow a user to set a power or temperature
threshold, and be notified when it is reached (or allow an action to
automatically be taken).
 Full CLI for all key AEM functionality. Support of AEM Server on AIX.
 Enterprise Integration and Manageability
 Out-of-the box management utilizing standard CIM profiles for AIX,
pLinux, i50S, HMC, and VIOS resources
90
© 2009 IBM Corporation
IBM Power Systems
WPAR Manager V2.1
Fully integrated as an IBM Systems Director plug-in
Support for Live Application Mobility with SAN Devices
91
© 2009 IBM Corporation
IBM Power Systems
Challenge: Virtualization Flexibility vs Complexity
Widespread enterprise
virtualization has led to
increased management
complexity
Physical server sprawl replaced
by virtual image sprawl
Multiple sysadmin skill sets
required to manage virtualized
infrastructure
Spending
(US$B)
Installed Base
(M Units)
$300
$250
$200
50
Power and cooling costs
45
Server mgmt and admin costs
40
New server spending
35
30
$150
25
20
$100
IT costs shifting from hardware
to energy, virtualization
infrastructure and management
15
10
$50
5
$0
0
Source: IDC, 2008
92
© 2009 IBM Corporation
IBM Power Systems
Solution: VMControl Unifies Virtualization Management
Enabling consistent multi-platform management for IBM Systems
 Manages Power Systems, System z®, System x®, storage and network resources
 Integrates management of virtual servers, appliances, storage, networks and clouds
Provides seamless integration into Tivoli enterprise service management solutions
93
© 2009 IBM Corporation
IBM Power Systems
System Pools: Next Step in Evolution of Virtualization
Managing a pool of server resources with single systems simplicity
Combines multiple virtual resources into one manageable entity
Automates virtual image mobility for optimal utilization and resilience
Optimizes virtual assets for performance, availability and energy use
Integrates server, storage and network virtualization
Mobility
Virtualization
Compute
Memory
Network
Storage
IT Resources
94
Compute
Memory
Network
Optimized for
Storage
Virtual Images
 Availability
 Performance
 Energy
System Pools
© 2009 IBM Corporation
IBM Power Systems
IBM delivers end-to-end visibility, control, and automation for
virtual machines and system images
 Orchestrate composite image processes
across the enterprise aligned with business
service requirements
 Manage virtual images - capture, configure,
catalog and deploy
 Standardize and centralize the
management of virtualization technologies to
gain operational efficiencies
 Manage virtualization across all IBM
Systems
z/VM®,
 Support for PowerVM™,
VMware,
Hyper-V and KVM virtualization
Optimize
•Automate business service
delivery
Manage
•Centralize enterprise image
management and deployment
Virtualize
•Create, manage and migrate
virtual machines (physical-tovirtual, virtual-to-virtual)
... reducing the time to deploy applications from days to minutes.
95
© 2009 IBM Corporation
IBM Power Systems
Systems Director VMControl
provides consistent, cross-platform management for virtualized environments
IBM Systems Director
Optimize
96
VMControl Enterprise Edition
Optimize With System Pools
 Create, modify, delete
 Automate resource mobility
 Manage utilization and availability
Manage
VMControl Standard Edition
Manage Virtual Image Libraries
 Create, capture, import, deploy
 Centralize image management
 Migrate virtual-to-virtual images
Virtualize
VMControl Express Edition
Virtualize Workloads
 Create, modify, delete VMs
 Manage multiple hypervisors
 Relocate VMs
© 2009 IBM Corporation
IBM Power Systems
VMControl Editions and PowerVM
VMControl
VMControl
Express Edition
Virtualization Capabilities
Virtualize resources
60-day Free Trial!
VMControl
Standard Edition
VMControl
Enterprise Edition
Manage virtual images Optimize system pools
PowerVM
Create/manage virtual machines
(x86, PowerVM and z/VM)



Virtual machine relocation



Capture/import, create/remove
standardized virtual images


Deploy standard virtual images


Maintain virtual images in a
centralized library


Create/remove system pools and
manage resources in system pools

Add/remove physical servers
within system pools

97
© 2009 IBM Corporation
IBM Power Systems
IBM Platform and Enterprise Service Management
Enterprise service management
Align IT operations with the business
Govern and control the business
Optimize the business
Platform management of IBM systems
Consolidated management across systems
Integrated physical and virtual management
Automated physical and virtual provisioning
The unique integration of IBM Systems Director and Tivoli® provides a
centralized platform for consolidated data center service management.
98
© 2009 IBM Corporation
IBM Power Systems
IBM Director and Tivoli
IBM Director
•System performance and
power measurement
99
Application
Middleware
IBM Director
Network
Operating System
Hardware
IBM Tivoli
•“Care and feeding” of the
hardware
•Detailed hardware
inventory, alerts and tools
for IBM Systems
•Basic patch management
•Allows for upward
integration into the Tivoli
environment
•Advanced, predictive
server hardware
management
IBM Tivoli
Resources
•Real-time ultra-scalable,
cross platform enterprise
service management
solutions
•Business service
management with robust
analytics, enhanced service
impact analysis and Key
Performance Indicators
•Sophisticated Layer 1-3
network management with
advanced root cause analysis
•Robust application discovery,
change and configuration
management and monitoring
•Advanced security operations
management, provisioning,
software distribution and
inventory
© 2009 IBM Corporation
IBM Power Systems
IBM Systems Director Active Energy Manager 3.1
Exploit Energy Scale capabilities in POWER6 processor-based servers
Power Trending
Thermal Trending
Effective CPU Trending
Power Savings
Power Capping
Support power savings for new POWER6 processor-based models
Discover and monitor legacy and select non-IBM systems through the
intelligent Power Distribution Unit (iPDU)
Display trending information per load group
Allows management of POWER6 legacy systems
Support low- to mid-range storage devices
Enhancements above PowerExecutive™ V2 (Windows®, xLinux)
Support for new x86 Systems
Cross-system monitoring and management support
iPDU support
System polling enhancements
AEM application supported on: Windows, x86 Linux, Linux on p
100
© 2009 IBM Corporation
IBM Power Systems
Active Energy Manager Monitoring Functions
“No Charge” Monitor Functions
Power Trending
Displays power usage for individual systems over time (in a graph or in table format) to
understand power usage trends within and across their systems
Thermal Trending
Displays information on the inlet and exhaust temperatures for individual systems one at a time to
understand thermal characteristics of systems so that temperature adjustments can be made
within the IT shop
iPDU (intelligent Power Distribution Units)
Enables support for power trending for older systems, low- and mid-range storage devices as well
as non-IBM systems. By plugging these systems into an intelligent PDU (a smart power strip)
AEM can collect power information from I/O drawers within the iPDU thereby giving a more
complete view of power used within a data center
Native Support
Extends power management functions such as power trending, thermal trending, and power
capping, originally available on System x™, to multiple IBM platforms enabling power
management functions on all IBM systems from a single console which reduces complexity
101
© 2009 IBM Corporation
IBM Power Systems
Active Energy Manager Management Functions
“Priced” Management Functions
Power Capping
Allocates a maximum power level a system can use without having to
worry about power usage above the maximum point
AEM will throttle the processor to use less power, which slows down the
server, if the system starts to consume more than the maximum level set
This feature can come into play if it gets too warm in the data center as
setting the cap will ensure that the system will not use more than that cap
value thus reducing power and thermal usage
Power Savings Mode
Enables a system to save up to 30% of normal CPU power usage
Power savings is enabled via an on/off switch which can be scheduled
during times of low utilization
Occurs automatically based on processor utilization if the function is
supported on the system
Allows management of power usage as work activity shifts across various
demands
102
© 2009 IBM Corporation
IBM Power Systems
Implementing Active Energy Manager
How does it work?
Hardware, firmware, and systems management software in
servers and blades can take inventory of components
No agents are required on the endpoint servers
Active Energy Manager totals up the power draw for each
server/blade and tracks that usage over time
When power is constrained, Active Energy Manager allows
power to be allocated on a server by server
Care taken that limiting power consumption does not affect performance
Sensors and alerts can warn the user if limiting power to this server could affect
performance
In the future group power policies may be developed across
groups of servers and reallocated dynamically based on past
history
103
© 2009 IBM Corporation
IBM Power Systems
IBM Software
104
© 2009 IBM Corporation
IBM Power Systems
PowerHA – Power High Availability
Leadership AIX High Availability and Disaster Recovery Product
An Industry-leading UNIX® High Availability and Disaster Recovery Product
Protect your critical business applications through reliable monitoring,
failure detection and automated recovery of business applications
Linux Support
SLES 9 and RHEL 4 and above support
Ease of Use Enhancements
Server
A
Workload
failover
Shared Disk
Server
B
Configure an PowerHA cluster or upgrade PowerHA on a
node without disrupting the target application
Fast Failover Detection through enhanced AIX
integration improves failover time
Recognize Application and Resource presence
Resource Dependency Graph
Adjustable preferences
PowerHA
105
© 2009 IBM Corporation
IBM Power Systems
Data Center versus Multi-Site HA/DR Solutions
Data Center Solution Strategy:





Deliver near continuous application service
Eliminate the affects of planned outages
Minimize unplanned outages
Plan for regular and sustained role swap operations
Primary focus is recovery time (RTO) & recovery point (RPO)
Offering: HACMP™ → PowerHA → now PowerHA System Mirror Standard Edition
Multi-Site Solution Strategy:




Recover operations at a remote location after a system or data center failure
Establish or strengthen a business recovery plan
Provide separate recovery location
Primary focus is recovery point (RPO)
Offering: HACMP XD → PowerHA XD → now PowerHA System Mirror Enterprise
Edition
106
© 2009 IBM Corporation
IBM Power Systems
Positioning PowerHA Technologies
Power Systems Availability Solutions
PowerHA
PowerHA SystemMirror
PowerHA SystemMirror for AIX
107
…….future products
PowerHA pureScale
PowerHA SystemMirror for i
* Planned 2010
PowerHA SystemMirror
for AIX Standard Edition
PowerHA SystemMirror for i Standard Edition
PowerHA SystemMirror
for AIX Enterprise Edition
PowerHA SystemMirror for i Enterprise Edition
© 2009 IBM Corporation
IBM Power Systems
Positioning PowerHA Technologies
Power Systems Availability Solutions
PowerHA SystemMirror
PowerHA pureScale
Active/Standby HA/DR Clustering Cluster Management/Interconnect
Data Center high availability
Technology
Multi site capability for disaster
recovery
PowerHA SystemMirror for AIX
PowerHA SystemMirror for IBM i
Low latency high performance data
transfer
Distributed cluster coordination
Centralized locking
Included with DB2 pureScale
108
© 2009 IBM Corporation
IBM Power Systems
PowerHA SystemMirror for AIX V6.1 Highlights
PowerHA SystemMirror Standard Edition targeted at
data center high availability solutions
PowerHA SystemMirror Enterprise Edition adds
support for multi-site high availability and disaster
recovery solutions
Enterprise Edition extends options for multi-site
storage resiliency
Today: IBM DS8000® & SVC, Metro Mirror &
Global Mirror
Now: Adding new support for EMC SRDF
GLVM configuration wizard
Easier to set up a cross site DR configuration
PowerHA with Dynamic LPAR
Automatically rebalances processor resources
after failover to partitioned backup system
109
© 2009 IBM Corporation
IBM Power Systems
PowerHA SystemMirror for AIX Editions
PowerHA SystemMirror for AIX
Standard
Edition
Enterprise
Edition
Datacenter cluster management


Application service HA stack


Shared storage management


Automated cluster validation


Integrated disk heartbeat


 New Editions packaging makes
multi-site DR solutions
more affordable
Fast disk failover


DLPAR management for HA


Cluster wide file symc


Smart Assists


 New tiered pricing structure
lowers costs of HA/DR
solutions
for mid-sized businesses
Multi Site HA Management

PowerHA GLVM async mode

Host based geographic mirroring

Host based deployment wizard

Supports IBM Metro Mirror

Supports IBM Global Mirror (SVC)

Supports EMC SRDF sync/async

 Standard Edition targeted
at datacenter HA
 Enterprise Edition targeted
at multi-site HA/DR
110
© 2009 IBM Corporation
IBM Power Systems
IBM Tivoli Access Manager
Defend against the top security threats
Protects against misuse by employees and
internal users
Prevents most hacking exposures
Increased security through fine-grained user
authorization
Secure control of all root user privileges
Comprehensive audit records of all root user
activity
Document regulation compliance
Available at no additional charge on Power
Systems
111
© 2009 IBM Corporation
IBM Power Systems
GPFS: High-performance, Highly Reliable File Access
 GPFS™ allows a cluster of nodes to read/write to a set of disks in parallel
at very high speed
 GPFS recently set a world record of 100 GBps sustained!
 Redundant data paths and extensive failure recovery ensure reliability
 Some commercial clients use GPFS entirely for its reliability




New data management capabilities increase efficiency and ease!
Multi-cluster capability enables data sharing across sites
Simultaneous file access eliminates the overhead of multiple copies
Powers many of the world’s most powerful supercomputers
 Now increasingly used in commercial applications
 Digital media
 Business intelligence
 Engineering design
 Medical imaging
 Geographic information systems
 Life sciences
 Data sharing
 Financial analysis
112
© 2009 IBM Corporation
IBM Power Systems
Business Change is Constant
 Can you support growing application workloads?
 Can you add extra capacity as you need it?
 Can you ensure business continuity at all times?
 Can you do it without changes to your applications?
 Can you do it efficiently and economically?
113
© 2009 IBM Corporation
IBM Power Systems
DB2 pureScale
Unlimited Capacity
Buy only what you need, add capacity as your
needs grow
Application Transparency
Avoid the risk and cost of
application changes
Continuous Availability
Deliver uninterrupted access to your data with
consistent performance
Leverages the architecture of z/OS:
the Gold Standard of reliability and scalability
Built on Power Systems and AIX
114
© 2009 IBM Corporation
IBM Power Systems
DB2 pureScale Architecture
Automatic workload balancing
Cluster of DB2 nodes
running on Power servers:
Power 550 or Power 595s
PowerHA pureScale
technology drives the
clustering
Runs on an LPAR or a
stand alone server
InfiniBand network
Shared Data – IBM Storage supported
115
© 2009 IBM Corporation
IBM Power Systems
Application Transparency
Avoid the risk and cost of application changes
Take advantage of extra capacity instantly
No need to modify your application code
No need to re-tune your infrastructure
 Run applications written for other database software
with little or no changes
Native support for commonly used
Oracle Database syntax and procedure language
DBAs can add capacity without re-tuning or re-testing
116
© 2009 IBM Corporation
IBM Power Systems
Unlimited Capacity: Scale with the Business Needs
 Designed to grow to whatever capacity your system requires
 Flexible licensing designed for minimizing costs of peak times
 Reduce cost with efficient use of system resources
Process higher transactional workload with minimal system overhead
Add/remove capacity as needed
Flexible software licensing to accommodate peak workloads
 Provide capacity for any transactional workload
In architecture validation testing, scales to more than one hundred DB2 servers*
Leverage highest performing database systems in the industry; over 6M tpmc per single system image**
Solution:
Use DB2 pureScale and
add another server for
those two days, and only
pay software license fees
for the days you use it.
* Validation testing includes capabilities to be available in future releases.
** TPC/C results. tpmC: 6,085,166; Price/tpmC: 2.81 USD; System Available: 12/10/08; Processors: 32; Cores: 64; Threads: 128.
DB2 pureScale helps CIOs handle business critical peak periods & save costs
117
© 2009 IBM Corporation
IBM Power Systems
Continuous Availability
Protect from infrastructure outage
Architected for no single point of failure
Redundant architecture
Continuous processing during node failure
All surviving nodes continue to process transactions during node failure
Data being modified by failing node is available in less than 20 seconds
Redistribute workload to available nodes immediately
Highly scalable and resilient infrastructure with IBM Power Systems
Exploits new PowerHA pureScale technology
Based on the most reliable UNIX platform according to ITIC*
Over 100+ node architecture validation has been run by IBM
118
© 2009 IBM Corporation
IBM Power Systems
PowerHA pureScale Technology Enables
Efficient and Continuous Operations
DB2 pureScale includes PowerHA pureScale technology to...
Reduce system overhead by minimizing inter-node communications
Centralized database locking and caching minimizes inter-node communications,
maximizing productive use of computing power
Reduce cost of systems communication with direct memory access
Remote Direct Memory Accesses virtually eliminates processor context switching
for IP network communications within the system
Maintain business continuity by minimizing impact of node failure
Data and lock status are immediately accessible to all nodes,
ensuring consistent application performance
Also exploits Power Systems 12X GX adapters to deliver
low latency & high performance interconnect
Power 595
Supported on Power 550 & Power 595
Power 550
Express
119
© 2009 IBM Corporation
IBM Power Systems
PowerHA pureScale Technology
Enables Efficient and Continuous Operations
DB2 pureScale includes PowerHA pureScale technology to...
Reduce system overhead by minimizing inter-node
communications
 Centralized database locking and caching minimizes inter-node communications maximizing productive use of computing power
Reduce cost of systems communication with direct memory
access
 Remote Direct Memory Accesses virtually eliminates processor context switching for
IP network communications within the system
Maintain business continuity by minimizing impact of node
failure
 Data and lock status are immediately accessible to all nodes,
ensuring consistent application performance
120
© 2009 IBM Corporation
IBM Power Systems
PowerHA Technologies
Power Systems Availability Solutions
PowerHA SystemMirror
Active/Standby HA/DR Clustering
Data Center high availability
Multi site capability for disaster recovery
PowerHA SystemMirror for AIX
PowerHA SystemMirror for IBM i
121
PowerHA pureScale
Cluster Management/Interconnect Technology
 Low latency high performance data transfer
 Global Lock management
 Global Buffer Pool
 Shared Communications
 Distributed cluster coordination
 Centralized locking
Included with DB2 pureScale
© 2009 IBM Corporation
IBM Power Systems
Achieving Efficient Scaling : Key Design Points
Deep RDMA exploitation over low latency
fabric
 Enables round-trip response time
~10-15 microseconds
Silent Invalidation
 Informs members of page updates requires no
CPU cycles on those members
 No interrupt or other message processing required
 Increasingly important as cluster grows
Buffers
Buffers
Buffers
Buffers
Hot pages available without disk I/O from
Global Buffer Pool memory
 RDMA and dedicated threads enable read page
operations in
~10s of microseconds
Global Buffer Pool
Global Lock
Manager
Shared Cache
122
© 2009 IBM Corporation
IBM Power Systems
Scalability : Example
Clients (2-way x345)
Transaction processing workload
modeling warehouse & ordering
process
 Write transactions rate to 20%
 Typical read/write ratio of many OLTP
workloads
No cluster awareness in the application




No affinity
No partitioning
No routing of transactions to members
Testing key DB2 pureScale design point
Configuration
 12 8-core p550 members

64 GB, 5 GHz each

64 GB, 5 GHz each

576 15K disks, Two 4Gb FC Switches

7874-024 IB Switch
 Duplexed PowerHA pureScale across 2
additional 8-core p550s
1Gb Ethernet
Client
p550
members
p550
powerHA pureScale
Connectivity
20Gb IB
pureScale
Interconnect
7874-024
Switch
 DS8300 storage
 IBM 20Gb/s IB HCAs
Two 4Gb FC
Switches
DS8300
Storage
123
© 2009 IBM Corporation
IBM Power Systems
Scalability : Example
12
11
10
9
8
7
6
5
4
3
2
1
0
Throughput vs 1 member
10.4x @ 12 members
7.6x @ 8 members
3.9x @ 4 members
1.98x @ 2 members
0
5
10
15
# Members
124
© 2009 IBM Corporation
IBM Power Systems
Power Systems Continues a 8+ Year Run of Growth
Sun SPARC and HP/Itanium continue to decline
UNIX® Server Rolling Four Quarter Average Revenue Share
Source: IDC Quarterly Server Tracker Q309 release, December 2009
125
© 2009 IBM Corporation
IBM Power Systems
What Drives IBM Power Systems Growth?
Power Systems leadership in Scale-up,
Scale-out and Scale-within
• Power Systems provide scalable,
predictable, consistent and balanced
system performance
Power Virtualization leadership
• For over 10 years, Power Systems has
been fine-tuning highly integrated
systems designed from the ground up
for industrial strength virtualization
SPARC, PA-RISC, Itanium and x86
users are moving to Power™
• Clients trust the migration experience of
IBM and the proven capability of Power
Systems to handle their toughest
workloads
126
© 2009 IBM Corporation
IBM Power Systems
Power Systems
Virtualization
Energy Efficiency
 60-80% utilization
 In 65% of systems shipped
in 2008
 70-90% energy cost
reduction
 More work per watt with
POWER6
High Performing, Scalable, Modular
127
Business Resiliency
Management
 Exploit PowerHA™
technologies
 Roadmap to continuous
 Increase deployment
speed
 Manage energy
usage
&
© 2009 IBM
Corporation
IBM Power Systems
Consolidating with AIX on Power Systems
enables clients to…
REDUCE COST
Server consolidation with shared resources enables high
system utilization, which lowers the cost of ownership
by reducing networking, energy, floor space, and
software costs.
IMPROVE SERVICE
Server consolidation improves service to clients by
delivering flexible performance, dynamic provisioning
and enabling clients to avoid disruption
MANAGE RISK
Server consolidation manages IT risk by improving
security, increasing business resiliency and
simplifying operations.
AIX, Power™ Systems and PowerVM™ are designed to deliver
effective consolidation in the most demanding data centers
128
© 2009 IBM Corporation
IBM Power Systems
Take The Steps to a Dynamic Infrastructure
with the New Power Equation
129
© 2009 IBM Corporation
IBM Power Systems
References
1. IBM AIX 6.1 Differences Guide SG24-7559
http://www.redbooks.ibm.com/abstracts/sg247559.html
2. AIX, VIOS and HMC Facts and Features
http://www.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/TD103130
3. IBM Infocenter
http://www-03.ibm.com/systems/p/support/index.html
4. Business Value of Power Systems
http://w3.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/PRS3655
http://cattail.cambridge.ibm.com/cattail/?source=v#[email protected]
5. IBM STG Sales Presentations
130
© 2009 IBM Corporation
IBM Power Systems
131
© 2009 IBM Corporation
IBM Power Systems
Special Notices
This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in
other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM
offerings available in your area.
Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions
on the capabilities of non-IBM products should be addressed to the suppliers of those products.
IBM may have patents or pending patent applications covering subject matter in this document. The furnishing of this document does not give
you any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY
10504-1785 USA.
All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives
only.
The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or
guarantees either expressed or implied.
All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the
results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations
and conditions.
IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions
worldwide to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment
type and options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal
without notice.
IBM is not responsible for printing errors in this document that result in pricing or information inaccuracies.
All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary.
IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply.
Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are
dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this
document may have been made on development-level systems. There is no guarantee these measurements will be the same on generallyavailable systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document
should verify the applicable data for their specific environment.
Revised September 26, 2006
132
© 2009 IBM Corporation
IBM Power Systems
Special Notices (Cont.)
The following terms are registered trademarks of International Business Machines Corporation in the United States and/or other countries: AIX, AIX/L, AIX/L (logo), AIX 6 (logo),
alphaWorks, AS/400, BladeCenter, Blue Gene, Blue Lightning, C Set++, CICS, CICS/6000, ClusterProven, CT/2, DataHub, DataJoiner, DB2, DEEP BLUE, developerWorks,
DirectTalk, Domino, DYNIX, DYNIX/ptx, e business (logo), e(logo)business, e(logo)server, Enterprise Storage Server, ESCON, FlashCopy, GDDM, i5/OS, i5/OS (logo), IBM,
IBM (logo), ibm.com, IBM Business Partner (logo), Informix, IntelliStation, IQ-Link, LANStreamer, LoadLeveler, Lotus, Lotus Notes, Lotusphere, Magstar, MediaStreamer, Micro
Channel, MQSeries, Net.Data, Netfinity, NetView, Network Station, Notes, NUMA-Q, OpenPower, Operating System/2, Operating System/400, OS/2, OS/390, OS/400, Parallel
Sysplex, PartnerLink, PartnerWorld, Passport Advantage, POWERparallel, Power PC 603, Power PC 604, PowerPC, PowerPC (logo), Predictive Failure Analysis, pSeries,
PTX, ptx/ADMIN, Quick Place, Rational, RETAIN, RISC System/6000, RS/6000, RT Personal Computer, S/390, Sametime, Scalable POWERparallel Systems, SecureWay,
Sequent, ServerProven, SpaceBall, System/390, The Engines of e-business, THINK, Tivoli, Tivoli (logo), Tivoli Management Environment, Tivoli Ready (logo), TME,
TotalStorage, TURBOWAYS, VisualAge, WebSphere, xSeries, z/OS, zSeries.
The following terms are trademarks of International Business Machines Corporation in the United States and/or other countries: Advanced Micro-Partitioning, AIX 5L, AIX
PVMe, AS/400e, Calibrated Vectored Cooling, Chiphopper, Chipkill, Cloudscape, DataPower, DB2 OLAP Server, DB2 Universal Database, DFDSM, DFSORT, DS4000,
DS6000, DS8000, e-business (logo), e-business on demand, EnergyScale, Enterprise Workload Manager, eServer, Express Middleware, Express Portfolio, Express Servers,
Express Servers and Storage, General Purpose File System, GigaProcessor, GPFS, HACMP, HACMP/6000, IBM Systems Director Active Energy Manager, IBM TotalStorage
Proven, IBMLink, IMS, Intelligent Miner, iSeries, Micro-Partitioning, NUMACenter, On Demand Business logo, POWER, PowerExecutive, PowerVM, PowerVM (logo), Power
Architecture, Power Everywhere, Power Family, POWER Hypervisor, Power PC, Power Systems, Power Systems (logo), Power Systems Software, Power Systems Software
(logo), PowerPC Architecture, PowerPC 603, PowerPC 603e, PowerPC 604, PowerPC 750, POWER2, POWER2 Architecture, POWER3, POWER4, POWER4+, POWER5,
POWER5+, POWER6, POWER6+, pure XML, Quickr, Redbooks, Sequent (logo), SequentLINK, Server Advantage, ServeRAID, Service Director, SmoothStart, SP, System i,
System i5, System p, System p5, System Storage, System z, System z9, S/390 Parallel Enterprise Server, Tivoli Enterprise, TME 10, TotalStorage Proven, Ultramedia,
VideoCharger, Virtualization Engine, Visualization Data Explorer, Workload Partitions Manager, X-Architecture, z/Architecture, z/9.
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, other countries or both.
Linux is a trademark of Linus Torvalds in the United States, other countries or both.
Microsoft, Windows, Windows NT and the Windows logo are registered trademarks of Microsoft Corporation in the United States, other countries or both.
Intel, Itanium, Pentium are registered trademarks and Xeon is a trademark of Intel Corporation or its subsidiaries in the United States, other countries or both.
AMD Opteron is a trademark of Advanced Micro Devices, Inc.
Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. in the United States, other countries or both.
TPC-C and TPC-H are trademarks of the Transaction Performance Processing Council (TPPC).
SPECint, SPECfp, SPECjbb, SPECweb, SPECjAppServer, SPEC OMP, SPECviewperf, SPECapc, SPEChpc, SPECjvm, SPECmail, SPECimap and SPECsfs are trademarks
of the Standard Performance Evaluation Corp (SPEC).
NetBench is a registered trademark of Ziff Davis Media in the United States, other countries or both.
AltiVec is a trademark of Freescale Semiconductor, Inc.
Cell Broadband Engine is a trademark of Sony Computer Entertainment Inc.
InfiniBand, InfiniBand Trade Association and the InfiniBand design marks are trademarks and/or service marks of the InfiniBand Trade Association.
Other company, product and service names may be trademarks or service marks of others.
Revised January 15, 2008
133
© 2009 IBM Corporation
IBM Power Systems
Notes on Benchmarks and Values
The benchmarks and values shown herein were derived using particular, well configured, development-level computer systems. Unless otherwise
indicated for a system, the values were derived using external cache, if external cache is supported on the system. Buyers should consult other
sources of information to evaluate the performance of systems they are considering buying and should consider conducting application oriented
testing. For additional information about the benchmarks, values and systems tested, contact your local IBM office or IBM authorized reseller or
access the following on the Web:
TPC
http://www.tpc.org
Linpack
http://www.netlib.no/netlib/benchmark/performance.ps
Pro/E
http://www.proe.com
SPEC
http://www.spec.org
GPC
http://www.spec.org/gpc
NotesBench Mail
http://www.notesbench.org
VolanoMark http://www.volano.com
STREAM
http://www.cs.virginia.edu/stream/
Unless otherwise indicated for a system, the performance benchmarks were conducted using AIX V4.3 or AIX . IBM C Set++ for AIX and IBM XL
FORTRAN for AIX with optimization were the compilers used in the benchmark tests. The preprocessors used in some benchmark tests include
KAP 3.2 for FORTRAN and KAP/C 1.4.2 from Kuck & Associates and VAST-2 v4.01X8 from Pacific-Sierra Research. The preprocessors were
purchased separately from these vendors. Other software packages like IBM ESSL for AIX and MASS for AIX were also used in some benchmarks.
The following SPEC and Linpack benchmarks reflect microprocessor, memory architecture, and compiler performance of the tested system (XX is
either 95 or 2000):
–SPECintXX - SPEC component-level benchmark that measures integer performance. Result is the geometric mean of eight tests comprising the
CINTXX benchmark suite. All of these are written in the C language. SPECint_baseXX is the result of the same tests as CINTXX with a
maximum of four compiler flags that must be used in all eight tests.
–SPECint_rateXX - Geometric average of the eight SPEC rates from the SPEC integer tests (CINTXX). SPECint_base_rateXX is the result of the
same tests as CINTXX with a maximum of four compiler flags that must be used in all eight tests.
–SPECfpXX - SPEC component-level benchmark that measures floating-point performance. Result is the geometric mean of ten tests, all written
in FORTRAN, included in the CFPXX benchmark suite. SPECfp_baseXX is the result of the same tests as CFPXX with a maximum of four
compiler flags that must be used in all ten tests.
–SPECfp_rateXX - Geometric average of the ten SPEC rates from SPEC floating-point tests (CFPXX). SPECfp_base_rateXX is the result of the
same tests as CFPXX with a maximum of four compiler flags that must be used in all ten tests.
–SPECweb96 - Maximum number of Hypertext Transfer Protocol (HTTP) operations per second achieved on the SPECweb96 benchmark without
significant degradation of response time. The Web server software is ZEUS v.1.1 from Zeus Technology Ltd.
–SPECweb99 - Number of conforming, simultaneous connections the Web server can support using a predefined workload. The SPECweb99 test
harness emulates clients sending the HTTP requests in the workload over slow Internet connections to the Web server. The Web server software
is Zeus from Zeus Technology Ltd.
–SPECweb99_SSL - Number of conforming, simultaneous SSL encryption/decryption connections the Web server can support using a predefined
workload. The Web server software is Zeus from Zeus Technology Ltd.
–SPEC OMP2001 - Measures performance based on OpenMP applications.
–SPECsfs97_R1 - Measures speed and request-handling capabilities of NFS (network file server) computers.
Revised September 24, 2003
134
© 2009 IBM Corporation
IBM Power Systems
Notes on Benchmarks and Values (Cont.)
–SPECjAppServer200X
(where X is 1 or 2) - Measures the performance of Java Enterprise Application Servers using a subset of J2EE APIs in a
complete end-to-end Web application.
The Linpack benchmark measures floating-point performance of a system.
–Linpack DP (Double Precision) - n=100 is the array size. The results are measured in megaflops (MFLOPS).
–Linpack SP (Single Precision) - n=100 is the array size. The results are measured in MFLOPS.
–Linpack TPP (Toward Peak Performance) - n=1,000 is the array size. The results are measured in MFLOPS.
–Linpack HPC (Highly Parallel Computing) - solves the largest system of linear equations possible. The results are measured in GFLOPS.
STREAM measures sustainable memory bandwidth in high performance computers.
VolanoMark is a 100% pure Java server benchmark that creates long-lasting network client connections in groups of 20 and measures how long it
takes for the clients to take turns broadcasting their messages to the group. The benchmark reports a score as the average number of messages
transferred by the server per second.
–The
following Transaction Processing Performance Council (TPC) benchmarks reflect the performance of the microprocessor, memory subsystem,
disk subsystem, and some portions of the network:
–tpmC - TPC Benchmark C throughput measured as the average number of transactions processed per minute during a valid TPC-C
configuration run of at least twenty minutes.
–$/tpmC - TPC Benchmark C price/performance ratio reflects the estimated five year total cost of ownership for system hardware, software, and
maintenance and is determined by dividing such estimated total cost by the tpmC for the system.
–QppH is the power metric of TPC-H and is based on a geometric mean of the 17 TPC-H queries, the insert test, and the delete test. It measures
the ability of the system to give a single user the best possible response time by harnessing all available resources. QppH is scaled based on
database size from 30GB to 10TB.
–QthH is the throughput metric of TPC-H and is a classical throughput measurement characterizing the ability of the system to support a multiuser
workload in a balanced way. A number of query users is chosen, each of which must execute the full set of 17 queries in a different order. In the
background, there is an update stream running a series of insert/delete operations. QthH is scaled based on the database size from 30GB to
10TB.
–$/QphH is the price/performance metric for the TPC-H benchmark where QphH is the geometric mean of QppH and QthH. The price is the fiveyear cost of ownership for the tested configuration and includes maintenance and software support.
Revised January 9, 2003
135
© 2009 IBM Corporation
IBM Power Systems
Notes on Benchmarks and Values (Cont.)
The following graphics benchmarks reflect the performance of the microprocessor, memory subsystem, and graphics adapter:
–SPECxpc results - Xmark93 is the weighted geometric mean of 447 tests executed in the x11perf suite and is an indicator of 2D graphics
performance in an X environment. Larger values indicate better performance.
–SPECplb results (graPHIGS) - PLBwire93 and PLBsurf93 are geometric means of literal and optimized Picture Level Benchmark (PLB)
tests for 3D wireframe and 3D surface tests, respectively. Larger values indicate better performance.
–SPECopc results - Viewperf 7 (3dsmax-01, DRV-08, DX-07, Light-05, ProE-01, UGS-01) and Viewperf 6.1.2 (AWadvs-04, DRV-07, DX06, Light-04, medMCAD-01, ProCDRS-03) are weighted geometric means of individual viewset metrics. Larger values indicate better
performance.
The following graphics benchmarks reflect the performance of the microprocessor, memory subsystem, graphics adapter and disk
subsystem.
–SPECapc Pro/Engineer 2000i2 results - PROE2000I2_2000370 was developed by the SPECapc committee to measure UNIX and
Windows workstations in a comparable real-world environment. Larger numbers indicate better performance.
The NotesBench Mail workload simulates users reading and sending mail. A simulated user will execute a prescribed set of functions 4 times
per hour and will generate mail traffic about every 90 minutes. Performance metrics are:
–NotesMark - transactions/minute (TPM).
–NotesBench users - number of client (user) sessions being simulated by the NotesBench workload.
–$/NotesMark - ratio of total system cost divided by the NotesMark (TPM) achieved on the Mail workload.
–$/User - ratio of total system cost divided by the number of client sessions successfully simulated for the NotesBench Mail workload
measured. Total system cost is the price of the server under test to the customer, including hardware, operating system, and Domino
Server licenses.
Application Benchmarks
–SAP - Benchmark overview information: http:// www.sap-ag.de/solutions/technology/bench.htm; Benchmark White Paper September,
2000;
– http://www.sap-ag.de/solutions/technology/pdf/50020428.pdf
–PeopleSoft - To get information on PeopleSoft benchmarks, contact PeopleSoft directly or the PeopleSoft/IBM International Competency
Center in San Mateo, CA.
–Oracle Applications - Benchmark overview information: http://www.oracle.com/apps_benchmark/
–Baan - The Baan benchmark demonstrates the scalability of Baan ERP solutions. The test results provide the number of Baan Reference
Users (BRUs) that can be supported on a specific system. BRU is a single on-line user or a batch unit workload. These metrics are
consistent with those used internally by both IBM and Baan to size systems. To get information on Baan benchmarks, contact Baan
directly or the IBM/Baan International Competency Center in San Mateo, CA.
Revised May 28, 2003
–J.D. Edwards Applications - Product overview information at http://www.jdedwards.com
136
© 2009 IBM Corporation
IBM Power Systems
Notes on Performance Estimates
rPerf
rPerf (Relative Performance) is an estimate of commercial processing performance relative to other pSeries
systems. It is derived from an IBM analytical model which uses characteristics from IBM internal workloads,
TPC and SPEC benchmarks. The rPerf model is not intended to represent any specific public benchmark
results and should not be reasonably used in that way. The model simulates some of the system operations
such as CPU, cache and maximum memory available. However, the model does not simulate disk or network
I/O operations. rPerf estimates are calculated based on systems with maximum memory and the latest levels of
AIX and other pertinent software. Actual performance will vary based on configuration details. The pSeries
640 is the baseline reference system and has a value of 1.0. Although rPerf may be used to compare estimated
IBM UNIX commercial processing performance, actual system performance may vary and is dependent upon
many factors including system hardware configuration and software design and configuration.
All performance estimates are provided "AS IS" and no warranties or guarantees are expressed or implied by
IBM. Buyers should consult other sources of information, including system benchmarks, to evaluate the
performance of a system they are considering buying. For additional information about rPerf, contact your local
IBM office or IBM authorized reseller.
Revised June 25, 2003
137
© 2009 IBM Corporation