AIX, VIOS and HMC Facts and Features

Ravi G Singh
IBM Power Systems rsingh@us.ibm.com

Release date: May 2010, latest updates in Red. Notice: This document is presented “As-Is” and IBM does not assume any responsibility for the
statements expressed herein. Any unannounced feature/product listed here is only a statement of direction by IBM and subject to change or withdrawal without notice. Author is not responsible for errors in this document that may result in any kind of inaccuracies.

Acknowledgement: Thanks to all the IBMers who contributed to this document and readers for their
valuable feedback and comments.

Updates: The latest document can be downloaded from one of the URLs below (depending on the
selected audiences). Internet: Intranet: BPs: http://www.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/TD103130 http://w3.ibm.com/support/techdocs/atsmastr.nsf/WebIndex/TD103130 http://partners.boulder.ibm.com/src/atsmastr.nsf/WebIndex/TD103130

Trademark and Copyright
IBM, the IBM logo, the e-business logo, System p™, System p5™, pSeries®, eServer™, OpenPower®, RS/6000®, POWER5™, POWER5+™, POWER6™, POWER™, PowerPC®, Power Architecture™, AIX®, AIX 5L™, BladeCenter®, IntelliStation®, EnergyScale™, HACMP, OS/2, PartnerWorld, pSeries, RS/6000, SecureWay, Tivoli and WebSphere are trademarks or registered trademarks of International Business Machines Corporation in the United States or other countries or both. A full list of U.S. trademarks owned by IBM can be found at: http://www.ibm.com/legal/copytrade.shtml. UNIX is a registered trademark of The Open Group in the United States and other countries. Microsoft, Windows and Windows NT are registered trademarks of the Microsoft Corporation. Java and all Java-based trademarks and logos are trademarks of Sun Microsystems, Inc. In the United States and/or other countries. Other company, product, and service names may be trademarks or service marks of others. Copyright
© Copyright IBM Corp. 2007, All Rights Reserved.

© Copyright IBM Corp. 2007

Page 2 of 40

AIX Facts and Features

Contents

AIX 5.x, 6.x and 7.x AIX 4.x VIOS HMC and Firmware HMC Ports HBA/FC Adapter Features Notes

……………………………………………………………………………… ……………………………………………………………………………… ……………………………………………………………………………… ……………………………………………………………………………… ……………………………………………………………………………… ……………………………………………………………………………… ………………………………………………………………………………

4 13 15 17 19 19 20

© Copyright IBM Corp. 2007

Page 3 of 40

AIX Facts and Features

AIX 5.x, 6.x and 7.x Facts and Features
Announcement/Offering General availability Withdrawn from marketing Withdrawn from support Latest Technology Level (TL) Tiers (Small, Medium, Large) Editions (EX, SE. & EE) – Note 50 Price included with H/W Hardware Support POWER7 POWER6 POWER5 POWER4 POWER3 PowerPC970 RS64, 604 Kernel Support UP & MP – Note 28 and 35 CHRP PCI and MCA Max. Processors HMT SMT2 SMT4 POWER7 Mode on P7 POWER6/6+ Mode on P7 CUoD Permanent Mobile - SOD CoD - Temporary Utility CoD Reserve CoD for shared proc pool On/Off CoD post pay Trial CoD Capacity BackUp (CBU) Vector Support (AltiVec) Decimal Floating Point Virtualization Micro-partitions max.-Note10 Virtualization – Processors, Memory, IO, Network and storage Dedicated Processor Sharing Multiple Shared processor pool Active Memory Sharing – Note 46 Active Memory Expansion–Note49 Virtual I/O server N N Y Y Y N N 5.1 N N N N N N N N N Y Y Y N N 5.2 N N N N N N N Y Y Y Y Y Y-P6 Y-P6 5.3 254-P5,P6,P7 Y-P5, P6, P7 Y-P6, P7 Y-P6, P7 N N Y Y N Y Y Y Y-P6 Y-P6 6.1 254-P5,P6, P7 Y-P5, P6, P7 Y-P6, P7 Y-P6, P7 Y-P6, P7 Y-P7 Y 7.x - SOD 1000-P7 Y N Y N Y N Y Y 5.1 2001-05-04 2005-04-29 2006-04-01 5100-09 N N Yes 5.1 N N N Y Y Y Y 32b, 64b UP & MP Y Y 32 Y N N N N 5.2 2002-10-18 2008-07-08 2009-04-30 5200-10 N N Yes 5.2 N Y-p570 Y Y Y Y Y 32b, 64b UP & MP Y N 32 Y N N N N 5.3 2004-08-20 2011-04 TBA 5300-12 S, M, L SE No 5.3 Y Y Y Y Y Y Y 32b, 64b MP Y N 64-P5, P6, P7 N Y-P5, P6, P7 N N Y 6.1 2007-11-06 TBA TBA 6100-05 S, M, L EX, SE, EE No 6.1 Y Y Y Y N Y N 64b MP Y N 64-P5, P6, P7 N Y-P5, P6, P7 Y-P7 Y Y Y Y-P7 Y N 256-P7 7.x - SOD Y Y Y Y N N N S, M, L EX, SE, EE 7.x - SOD TBA TBA TBA

© Copyright IBM Corp. 2007

Page 4 of 40

AIX Facts and Features

iSCSI MPIO DMP-IO SCSI Virtual I/O and Ethernet FCoE or FCoCEE or CNA I/O Drawer Dynamic Configuration USB Drive support SSD Support 6.2 X11R6. Virtual processors/LPAR Note 19 Min.x . memory/LPAR in MB-Note12 Memory incr.SOD N N N N N N N N N N N N Y-ML F N N N N N N N N N Y-P5 Y Y-P5 Y-P6. P7 Y-P7 LPAR DLPAR-Processors.2 WPAR Manager Live Applicatiion Mobility – Note 47 Live Partition Mobility Partition Hibernation . P7 Y-P6. P7 Y-P6.4 7.Integrated Virtualization Manager (IVM) – Note 24 Partition Load Manager (PLM) Virtual Ethernet – Note 13 Integrated Virtual Ethernet (IVE) – Note 38 NPIV .1 N N 32 N 256 256 N 5.2 Y N 64 N 256 256 N 5.2 N Y-p570 Y Y Y-ML03 Y Y N N N N N 5. P7 No N N N Y-P6. P7 Y>=P4 N Y Y>=P4 Y-P6.1 N 5. P7 Y N Y Y-P6.3 X11R6.3 Y-64b Y Y Y Y Y Y Y Y Y Y Y 6.SOD © Copyright IBM Corp.8Gb PCIe HBA WPARs (Max 8192 active) WPAR with AIX 5. Memory and I/O slots Multiple Instances of AIX on a Single rootvg – Note23 Maximum LPARs – Note10 Max.SOD I/O Infiniband (PCI & GX) – Note 22 PCIe PCI-X High Performance Switch (HPS) IP over FC.1 X11R7.1 Y-V2.4 6.1 X11R6.1 N N Y N N N N N N N N N 5.x .1 Y-V2.1 Y Y Y Y Y Y Y Y Y Y 7.1 Y Y 254 64 128 16 7.1 N 5.x .SOD X11/CDE X Release GNOME Desktop 5. 2007 Page 5 of 40 AIX Facts and Features ./decrement in MB Application & LPAR Accounting 5.3 Y Y 254 64 128 16 Y 5.

P7 Y Y Y Y N Y Y N Unlimited 3.SOD Permanent Unlimited 2 10 Permanent Unlimited 3.25 13 System Management & RAS WLM. CT®.TOD Management Multiple instances of AIX in one rootvg (multibos) Thin Server & Common OS Image (COSI) – Note30 NIM GUI Installer Dump analysis Live System Dump – Note 34 Mini Dump Parallel Dump Component Dump & dumpctrl FW assisted dump System dump on reset DVD Support for system dump Probevue System hang detection Dynamic Processor deallocation Dynamic Processor sparing Kernel Recovery (4KB stack/ thread. P7 Y Y Y 7. 2007 Page 6 of 40 AIX Facts and Features .CPU. heap. 64KB. P5.P6.2 1024 16TB Y N 5. I/O .MEMORY Max.1 N N N N Y N Y N 5. Memory.25-32b Pbytes-64b 13-32b 64M-64b 5.LMT.3 2048-64b 96-32b 16TB 64b-Dynamic N 6.x .1 4096 32TB Dynamic Y-P6 7.SOD N N Y N Y Y Y N Y Y Permanent Unlimited 3. Real Server Memory in GBNote10 Max Memory supported 4KB.x .3 Y Y Y Y Y N Y Y Y N N Y Y N Y Y Y N N Y-P5 . 64KB stack/processor Concurrent AIX Updates – eFix Concurrent Firmware upgrade EEH FFDC enhancements . P6 Reserve CoD On/Off CoD post pay – P5.1 Y Y Y Y Y+NFS4 Y Y Y Y Y>=P6 Y Y Y Y Y Y Y Y Y Y-P5.25-32b Pbytes-64b 13-32b 64M-64b 6.2 Y Y N N Y N Y N N N Y N N N N N Y N N N N Y Y Y N N Y-P5 Y LMT N © Copyright IBM Corp. stack and shared memory CUoD (Permanent) – P4.1 512 16TB N N 5. P6 Capacity BackUp (CBU) Address Space segments Data heap Segment size in GB – Note 33 Program size in 256MB segments 5.P6. and RTEC – Note 37 FFDC OS 5. P5. P6 Trial CoD) – P4. 16MB & 16GB Memory page – Note 26 Dynamic Variable page size (4K to 64K) for data.

9.11.11.0 Y Y V3 V2. of EC per system Network Data Administration Facility (NDAF) TLS (Secure ftpd RFC 4217) RFC1323 enabled by default – Note 42 Default biod daemons IGMPv3 RFC3376 – Note39 IP over Fibre Channel 5. of ENT adapters per EC Max no. 4. 9. 8. and krb5p) NFS-statd Threading AutoFs Network Caching daemon netcd CIO & DIO for client NFS Proxy Serving with Caching Max for thewall Max.x .02.8.SOD © Copyright IBM Corp.3 Y N V3 V2 & 3 N Mutli Mutli N N N Note 6 1024 interface Y 8 8 N N N 6 N Y 5.0. krb5i. of interfaces for ENT isno tuning level Etherchannel (EC) Dynamic Adapter membership – Note 44 Max no.02.0. VIPA SLP V2 and SCTP – Note 3 SNMP NFS NFSV3 with Kerberos (krb5. 8.9.3 Y N V1 V2 & 3 N Mutli Mutli N N N Note 6 1024 interface N 8 8 N N N 6 N N 5. V4 Y Note 6 No limit interface Y 8 256 Y Y Y 32 Y Y 7.02.3 & 4 Y Mutli Mutli Y Y-V3. no.1. 2007 Page 7 of 40 AIX Facts and Features .8. 8.0 8.0 9.8.14 9.8 Y Y V3 V2.6p2 9.11. 4.0.Processor instruction retry Alternate Processor recovery RIO/GX Adapter Hot add and Cold repair – Power 570 and 595 Hot Node add/Cold node repair Note 43 Hot node repair/Hot memory add Note 43 PCI adapter hot swap Dynamic bit-steering (Sparing) Memory page deallocation Memory/Storage Keys Application Memory/Storage Keys Kernel N N N N N Y N N N N Y>=P6 Y>=P6 N N N Y N N N N Y>=P6 Y>=P6 Y>=P6 Y>=P6 Y>=P6 Y Static-P5 Y Y>=P6 N Y>=P6 Y>=P6 Y>=P6 Y>=P6 Y>=P6 Y Y Y Y>=P6 Y>=P6 TCP/IP and networking IP IPv6 RFC 4007 and RFC 4443 DHCP Sendmail BIND Multipath routing and DGD.0.1 V4/V6 N 8.1 V4/V6 Y V4/V6 8.14 8.4.3 V4/V6 N V4/V6 8. V4 Y Note 6 No limit interface Y 8 256 Y N N 6 N Y 6.3 & 4 Y Mutli Mutli N Y-V3.13.13.2 V4/V6 N 8.

1-2001 PAM Kerberos POSIX Realtime API No.512 Note 5 128 2048K Y Variable Both ioo Both Y 32MB 7.1 N N 5.Security & Compliance CAPP/EAL4+ .4.1 LVM Max PVs in a VG Max LVs in a VG – Note 5 Max LPs/LV Max PPs/disk PP Size in GB Max PPs/VG Hot spare disk in a VG Logical Track Grp Size (LTG) KB AIO – Legacy and POSIX command to tune Paging space Increase/Decrease Paging space garbage collection . of logged users max User/group name length (chars ) Max.x .Note31 UNIX03 UNIX98 & C99 ISO/IEC 9899:1999 IEEE 1003.2 128 512 32.Note 40 RBAC Stack Execution Disable (SED) Trusted AIX AIXPert SOX-COBIT compliance Trusted Execution (TE) Secure by Default (SbD) File Permissions Manager (fpm) Encryption Support Network Authentication Service RFC 1510 RealSecure Server Sensor IBM Compliance Expert 5.SOD V5 N 8 16000 8 N N N N N N N N N PKCS11.2 Y Y Y-ML04 Y Y Y Y V5 5.1 128 512 32. Crypto Lib Y-1. Group line length (chars ) Password length (chrs) Loadable Password Algorithm (LPA) . CCS N N N Y-ML03 8 16000 8 N N N N N N N N Y PKCS11. Crypto Lib N Y 1.7 Y 1.3 1024 4095 32.3 Y Y Y Y Y Y V5 Y 32.Note 41 Boot image Size – Note 27 5.0. 2007 Page 8 of 40 AIX Facts and Features .Note31 LSLPP/EAL4+ .512 Note 5 1 130.1 Y Y 7.767 255 16000 Upto 255 Y N Y N Y N N N Y PKCS11.x .048 Y 128-1024 Legacy lsattr Incr only N 12MB 5.SOD © Copyright IBM Corp.1 6.1 1024 4095 32. CCS N N N 255 16000 Upto 255 Y Y Y Y Y+LDAP Y Y Y Y PKCS11.048 Y 128-1024 Both lsattr Both N 12MB 5.512 Note 5 128 2048K Y Variable Both aioo Both Y 32MB 6.512 Note 5 1 130.

32TB J2-1PB J-64GB J2-16TB J-64GB J2-1TB J-64GB 3.SOD © Copyright IBM Corp.SOD J2-4PB J2-4PB J-16TB/64b J-1TB/32b J2-16TB J-1TB J2-1PB J-64GB J2.072 268435456 11 N 15 Y Y Y N 5.FILE SYSTEM Filesystem Architectural Max 5.x . 64b process Max Shared memory IDs Max segments/process 64b Max segments/process 32b 32MB N 15 Y N N N 5.288 2GB 1TB 131.767 4MB 4MB 131. 766 Dynamic N Both N N >2 Dynamic 32b-1GB 64b-64GB 256KB to 16GB N 15 Y Y N N 5.728 65.1 J2-4PB 7. 766 Dynamic N Both Y Y >2 Dynamic Filesystem Max tested – Note 20 File Size Architectural Max File size Max tested File Descriptors or open files File Descriptors/process Max.1 128K-32b 1024K-64b 65.767 4MB 4MB 128K-32b 1024K-64b 524.072 65.728 65.288 2GB 64GB 131.145. ID Semaphore Max value MESSAGE QUEUES Max Message size Max bytes on queue Max message queue IDs Max messages per queue ID SHARED MEMORY Max Segment size . 2007 Page 9 of 40 AIX Facts and Features . sub directories/dir–J & J2 Max file system i-nodes for j2 J to J2 conversion on migration JFS and JFS2 – Note 20 JFS2 Quota JFS2 Shrink RAM disk or cachefs in GB Outline JFS2log Max J2-1TB J-1TB J2.2 131. 766 Dynamic N Both N N 2 1GB J2-16TB J-64GB 3. 32b process Max Segment size .728 32.288 2GB 1TB-32b 32TB-64b 128K-32b 1024K-64b 268435456 11 7.535 32.535 32.288 2GB 1TB-32b 32TB-64b 128K-32b 1024K-64b 268435456 11 Y-J2 64 Y Y Y-J2 6.2 5.145.145.072 524.072 524.767 4MB 4MB 131.072 65.3 6.1 131.534 32.072 268435456 11 Dynamic Y Both Y Y >2 Dynamic Inline JFSl2og Max Disable JFS2 logging Snapshot generations CDROM FS DVDRAM FS Freeze and Thaw – Note 29 Encrypted File system (EFS) SEMAPHORES Max Semaphore IDs Max Semaphores per semph.3 128K-32b 1024K-64b 65.32TB J2-1PB J-64GB J2-16TB J-64GB 3.767 4MB 4MB 128K-32b 1024K-64b 524.1 4PB 5.535 32.535 32.x .534 32.768 32.

303 2^31-1 N N N N N System N 1 10 4096 0/0 0/0 20/80/80 1 0 Y N N N N N N N System N 1 10 4096 0/0 0/0 20/80/80 1 0 N N N Y Y N Y Y Filesystem Y 3/core 30/core 65536 1/1 4096/8192 03/90/90 0 1 Y Y Y Y © Copyright IBM Corp.3 32.2 32.768 Y 7. AIO and Oracle – Note36 Solution Performance Tuning – Note 36 I/O Pacing tuning level I/O pacing enabled by default aio_minservers aio_maxservers aio_maxrequests aio_fastpath & aio_fsfastpath minpout/maxpout minperm/maxperm/maxclient lru_file_repage page_steal_method Memory Affinity (To disable vmo -o memory_affinity=0) nmon integration into topas VIOS monitoring in topas 5.1 32.000 5K P 8:1 ksh93/ ksh88 S 1:1 ksh93/ ksh88 25K Y-P6 N 2^31-1 2^31-1 2^31-1 2^31-1 2^31-1 2^31-1 2^31-1 N Y N N N Filesystem N 1 10/core 4096 0/0 0/0 20/80/80 1 0 Y Y Y 25K Y-P6 Y-P6 N N N N 2^31-1 4.767 N 5.767 N 6.SOD P 8:1 ksh93/ ksh88 P 8:1 ksh93/ ksh88 256.x .264 2^31-1 2^31-1 4. no.194. 2007 Page 10 of 40 AIX Facts and Features .KERNEL/PROCESS/TUNING Max Threads/process Limit max Threads/process and max processes/user (chuser and ulimit commands) AIXTHREAD_SCOPE – Note 45 AIXTHREAD_MNRATIO– Note 45 ksh Max Processes/system Max.303 523.194.767 N 5.1 32. of devices – Note 32 Memory/Storage Keys Application Memory/Storage Keys Kernel Resources (/etc/security/limits) CPU time File size Stack Size Memory Size Open Files Data Size Core Size Restricted tuneables – Note 34 AIX Run Time Expert AIX Profile Manager Out of the box Perf Tuning VMM.

1 6.1-P6 5.1.5.1.0 8.3 1. 7.4. 1.2 3. 7.0 2. 8.2-P6 © Copyright IBM Corp.Licensed Software and Tools XL C/C++ XL C XL Fortran 5. 8.0 11. 1.3. 9.3 N 6.0. 5.5.2 6.1-P6 1.1.2 6.1-32b 5.1 N 6.0 10.1 3.0.1 3.1.1. 6.2 2.2.2.0 3. 7. 5.1.3 3. 10.2.1 13.1.1 N Y Y Y 6. 5. 5.4.2. 7.1 N >=1.1.1.7-P6 3.4 or 3. 4. 1. 5.1 11.2 6.2 2. 12.2.0 2.3 3.1 N 6.4. 5.4.1 5.5 5.0.2 6. 11.1.6.1.2. 3.0.6.0 6.2 3. 8.0.0 9. 8.5 5.0 2.4.3 2.1.1.0/5.1.3 11. 7. 5.1 Y 4.1.2. 10.0. 2.1. 5. 5.3 1. 9.1. 1. 6.4.5 3.2 6.2 Y Y Y Y 6. 1.1. 9.5.0 13.1.1 5.0. 3.2.1 6. 1. 5. 3.SOD VisualAge C++ C for AIX COBOL PL/1 AIXlink/X.1 2.0/5.1 11.1.4.3 eWLM Server Agent IBM Systems Director Server Common Agent Platform Agent IBM Director Server IBM Director Console IBM Director Agent IBM Systems Director Console VMControl – Note 48 Network Control .2.1.0. 5. 9.0.1.0.3.2.1 3.3 3.1.0.1 6.2.1.2.2 3.3. 2. 1. 5. 10.2. 7.1 2.2.6. 1.0.0/5. 5.1 6. 5.0. 10.1. 8.1 3. 10.5.3 3. 5. 21 3.0 11. 2. 2.1 6.3.1-P6 N 6.1 7.2 6.1 N 5.1.1 2.0/5. 4.0.1.1 N N N N 3.1. 1.2.3.1.0 2.3 N N N 5.3.0. 5.1.1 6.1. 2.0.5.0.0. 4.2 9. 12.5.1.0. 1. N N 1.1 6.x .3.25 AIX Fast Connect Communication Server DCE DFS CSM-Mgmt.2. 8.1. 8.1.Note 48 Active Energy Manager (AEM) – Note 48 Electronic Service Agent PM for Power Systems agent Tivoli Workload Scheduler (TWS) LoadLeveler POE 6.1 6. 7.1 8. 9. 7.1 6. Server – Note25 CSM-managed Nodes – Note 25 GPFS Ver Max Filesystem size – Note 7 PSSP – Note17 PowerHA (HACMP) N N 3.1-32b 1.3. 10. 1.1 3.0 2.3.4.3 Y 4.5 N 2. 5.0 2. 5. 3.1. 5.3 N 5.1.1 6.1.0. 2007 Page 11 of 40 AIX Facts and Features . 1.1.6.2.1.5.0.2.2-P6 N N N N N N N 3.0.1.0.

0.4 5.7.5302 4.9 N Y 6.8.3.11.4.53.8/7.0.0.1 2.1 2. 142.Expansion pack/Web download – Note9 RSCT OpenSSH 5.4 4. 1.0.0 N N Y Y 0.5.1 7.7.2.0.0.53 7.53.20 1.0 2.2 2.11 1.53 6.0.8.5.3 5. 5.3 2. 1.1.0.7.0.5.8.0 2.1 5.9.1 5.53 2. 131 © Copyright IBM Corp.0.3.0.53 4. 4.9. 2.19.52. 1.20 1.0.0.1 7.13.12 6.0.0.4 4.4.9.0.SOD OpenSSL lsof IBM HTTP Server 128-bit Netscape Communicator 128-b Mozilla FireFox 1.7.6.47.8. 142 2.53.7.2 N 5.2.52 4. 1. 5. 1.8.11.8 4.3. 131 5.6. 1.0.8.47.5.5.8.4.8. 0.1.2.4.8. 5. 3.1 5.7.52 5. 142.0. 3.0.8 2.0.0 2. 5.2.0.8.8.2 Cryptographic Library Perl Samba Adobe Reader Documentation Library AIX Information Center JDK Supported Version/Release 142.5302 4.11 1.0.3.13.26.2.0.x .8 4.5.0.0.4.0 4.2-256-bit 5.12 2. 2007 Page 12 of 40 AIX Facts and Features .7. 131 NA 5.7.52.19.6.3.7.0.06 Y Y 6.0.6.10.0.47.9.53 0.2 5.3.1.0 N N Y Y 2.0 5. 1. 3.5.0.13.2 5.1 NA 3.5.2. 1.9. 2.0.0.0.0. 0.53 4.8.2 3.53.

000 32.767 2^32 256MBx16 2GB 10 2^32 256MBx16 2GB 10 2^32 256MBx16 2GB 10 2GB 4GB 4GB 16GB 7*2^56 7*2^56 2^64 256MBx16 2GB 10 256MBx16 2GB 10 256MBx16 2GB 10 32GB 96GB 256MB 256MB 256MB 256MB 256MB JFS JFS 200.x Facts and Features Announcement/Offering General availability Withdrawn from marketing Support withdrawn Latest Maintenance Level (ML) Hardware Support CHRP PCI. MCA LVM Max PVs in a VG Max LVs in a VG Max PPs/disk Max PPs/LV Max Major nos.000 2.3 17-Sep-99 30-Jun-03 31-Dec-03 4330-11 © Copyright IBM Corp.000 200.1.3.767 4.767 4.535 32.535 32.2.096 65.000 200.535 32.4/5 25-Apr-97 31-Dec-98 31-Mar-99 4.535 32.096 65.767 2GB 64GB 64GB 64GB 64GB 64GB 128GB 1TB 1TB 1TB 1TB 64GB 64GB 1.072 65. Data & BSS in Bytes Symbol Values (address) Address Space segments Data heap Segment size Program size in 256MB segments SEMAPHORES Max Semaphore IDs Max Semaphores per ID Semaphore Max value 4. Max VMM Mapped devices Hot spare disk in a VG Logical Track Grp Size (LTG) KB AIO Paging space Increase/Decrease FILE SYSTEM Filesystem Architectural Max Filesystem Max tested File Size Architectural Max File size Max tested File Descriptors or open files File Descriptors/process Max file system inodes Max file system fragments JFS and JFS2 RAM disk or cachefs in GB Outline JFSlog Max Inline JFSlog Max MEMORY Real Memory Size of Executable in Bytes Size of Text.768 Legacy Incr only Legacy Incr only Legacy Incr only Legacy Incr only Legacy Incr only 256 512 256 512 256 512 256 512 256 512 No 128 Legacy Incr only 32 256 1016 32 256 1016 32 256 Note 5 32 256 Note 5 128 512 Note 5 128 512 Note 5 Y Y Y Y Y Y Y Y Y Y Y Y 4.2 23-Oct-98 Sep-99 31-Dec-01 4.1 25-Apr-97 31-Dec-99 31-Mar-00 31-Dec-01 4.048.2.000 2.000 2.AIX 4.767 2^24 2^28 JFS JFS JFS 2 1GB 32MB 1.576 32.0 4. 2007 Page 13 of 40 AIX Facts and Features .048.767 131.576 32.3.535 32.3.072 65.000 2^24 2^28 JFS 1000.096 65.767 4.767 131.096 65.535 32.1 24-Apr-98 4.

8.767 512.535 4. 122.1 No V1 No V1 Yes Yes Yes Yes Yes 131.192 256MB NA 4.288 2GB 2GB 131.096 10 1 512 262.535 65.072 524.767 262.1.538 11 1 32.3 No V1 Single X11R6.072 32 Note6 ksh88 core V4/V6 8.3.3.192 256MB NA 4.072 32 64MB core V4 4.143 131.3.192 256MB NA 4.75 6.1 - Licensed Software Support VisualAge C++ C for AIX XL Fortran AIXlink/X.072 131.2.1.8 8.1 65.Announcement/Offering MESSAGE QUEUES Max Message size Max bytes on queue Max message queue IDs Max messages per queue ID SHARED MEMORY Max Segment size .535 4.1 No V1 4.3 4MB 4MB 131.7 4.05 1.1.192 2GB 2GB 65.288 2GB 2GB 131.535 65.535 4.3.072 32 128MB ksh88 core V4/V6 8.6 8.8 8.096 8.538 131.79 1.1 65.096 10 1 512 65.767 131. Memory I/O Bonus and Expansion Pack Adobe Acrobat Reader IBM HTTP Server 128-bit Netscape Communicator 128-b JDK Cryptographic Library 4.9.1 3. 118 Y N 4. 2007 Page 14 of 40 AIX Facts and Features .8 No V1 4.1/7.9.000 32 Note6 ksh88 core V4/V6 8. segment size KERNEL/PROCESS Max Kernel Threads/process Max Threads/System Max processes/user Max Processes/system Max concurrent groups/process Max for thewall ksh Core File naming TCP/IP and networking IP Sendmail BIND Multipath routing and DGD.072 32 64MB ksh88 core V4 8.7 4.2 4MB 4MB 131.5.1 5.535 65.072 524.072 131.12.25 AIX Fast Connect Communication Server DCE © Copyright IBM Corp.1 6.072 32 64MB core V4 8. VIPA SNMP NFS-statd Threading & AutoFs X11 X Release Security C2 JDK Supported Version/Release New Features/Enhancements WLM.4 4.8.096 8.096 10 1 512 65.4/5 65. 130.535 4.096 8.CPU.3.072 131.0/5.072 11 1 32.143 131. 32-bit Max Segment size .0/3.2.000 170. 64-bit Max Shared memory IDs Max segments/process 64b Max segments/process Min.535 65.3.096 8.0 65.072 268435456 11 1 32.3 No V1 4.1.538 131.

3 Y N N Y Y N Y 5.09 V1.4 7.3 5.10 N N N N N N Y N N N N N N N N Y N N N N N N N N Y N N Y N N N N N Y N N Y N N N N N Y Y Y Y Y Y Y Y-P6.8Gb PCIe HBA Virtual Tape .3 Y Y Y Y Y N Y 6.2 2005.09 V1. 2007 Page 15 of 40 AIX Facts and Features .3.IVE.1 2008.3 2006.10 2008.1 Y Y Y Y Y Y Y Y Y N N Y N N 5.7.SAS Virtual Ethernet Network bandwidth apportioning for SEA IPV6 Y Y Y Y N Y N N Y N N Y N N Y Y Y Y N Y N N Y Y N Y N N Y Y Y Y N Y Y N Y Y N Y N N Y Y Y-P6 Y-P6 N Y Y N Y Y N Y N Y Y Y Y-P6 Y-P6 N Y Y Y Y Y Y Y Y Y Y Y Y-P6 Y-P6 Y-P6.3 N N N N Y N Y 5. NetApp. AIX6.3 Y Y 7.09 V1.11 V2.4 2007. iSCSI and Fibre Channel attach SAS and SATA SSD Support NPIV over FCoE CNA/FCoE or FCoCEE NPIV .1 >=5.1 2005. 64 Ports Y Y Y Y © Copyright IBM Corp.09 V1.3 >=5.x-P7 V1.3 5.07 2009.06 2010.3 Y N N N Y N Y 5.3 >=5.VIOS Facts and Features Features General availability End of Support Hardware/OS Support P7 P6 P5 P4.5 2007.+ Dedicated I/O + New features Shared Ethernet SEA Failover VSCSI disk image (cpbd) VSCSI for LUN and LV VSCSI for Filesystem Storage Support nSeries.3 5. P3 Power Blades AIX build Client OS Support .3 Y N N Y Y N Y 5.04 2008.AIX Linux IBM i HMC level required Virtualization Dedicated LPAR Shared Processor LPAR Multiple Shared Processor pools Shared Dedicated processor Capacity Active Memory Sharing Note 46 Integrated Virtualization Manager – Note 38 IVM – DLPAR (Proc & Memory) IVM .

svmon.1 N N N V1.1 N N N N N N N 10 N N N N N V1. viostat. ITUAM.com/webapp/set2/sas/f/power5cm/power6.3 N N Y V1.1 Y Y Y N N N N Y Y N N N Y Y Y IBM Tivoli Integration: IBM Tivoli Monitoring IBM Tivoli Storage Manager IBM Tivoli Usage and Accounting Manager IBM Tivoli Application Dependency Discovery Manager IBM Tivoli Productivity Center VIOS URL http://www14. 2007 Page 16 of 40 AIX Facts and Features . fcstat.4 N N Y V1.5 N N Y V2.software.2 N N N N N N N 10 N N N N N V1.2 N N N V1. ITM.Features GVRP (GARP VLAN Registration Protocol) Live Partition Mobility (LPM) Features Multiple HMC Support Preserve Virtual Target Device (VTD) Encrypt data during LPM Mover Service Partition Non Symmetric VIOS Maximum VIOS LDAP SNMP SPT CIM (Common Inf.1 Y Y Y Y Y Y Y 10 Y Y Y Y Y Features System Management AIX Runtime Expert Performance Manager Performance Management topas.4 Y Y N N N N N 10 Y Y Y Y N V1. Model) Expansion Pack V1.software.com/webapp/set2/sas/f/vios/documentation/installreadme.5 Y Y N N N N N 10 Y Y Y Y Y V2.html © Copyright IBM Corp. TPC – Tivoli Productivity Center Browser based HMC GUI V1.ibm. vmstat IBM Tivoli Integration TSM.html Power Code Matrix http://www14.3 N N N N N N N 10 N N N N N V1. TADDM.ibm.

0 is supported on IBM Power Systems with POWER7. P7 7042-C07 for P5. P7 7042-CR4 for P5. POWER6 or POWER5 processors.html • HMC V7. and 7042-C06.16 R7.08. Environments may include 7310-C03.1 2006.Opt N N N N N N N N Y Y N Y Y Y Y Y Y Y Y Y N Y N N N N N N N N N Y Y Y N Y V3R3. Each console must be configured with feature number 0962. • POWER7 Support: HMC V7 R710 is the minimum level for POWER7 support. -C05.1.10 R3. P6.07.software. P6.ibm. P6. P7 7310-C05 for P5.5. Letter 109-577). managed systems Max 590/595 Managed Max. P7 7310-C06 for P5.software. the servers can be attached to two separate HMCs to address availability requirements. http://www14.507 HMC Support URLs https://www14.HMC and Firmware Facts & Features HMC Code General availability Latest Maintenance Level (ML) Max.CR5 HMCs. P7 7310-CR4 for P5. as well as 7042.06 R5.com/webapp/set2/sas/f/power5cm/supportedcode.html http://www14.506 FF>= 1. P6.12 48 32 254 V5R2. LPARs – Note18 Hardware Support 7042-C08 for P5. P7 7042-CR5 for P5.09.19 48 32 254 V6R1.10.x 2005.06 48 32 254 V5R1.x 2006. -CR4. All HMCs must have the same level of HMC Licensed Machine Code (Ref 7042-CR5 Annc. -CR3.01.06.x 2010.software.0 2009.ibm.3 2007.Opt Y Y Y – Opt Y – Opt Y – Opt Y – Opt Y – Opt Y .7.10 48 32 254 V7R7.31 32 32 64 for 7315 V4R5. must be a CR3 or later model rack-mount HMC or C05 or later deskside HMC.7 2006. P6.html Notes on HMCs • If redundant HMC function is desired.1.ibm.ibm. -CR4.software. -C06.28 R2.02 48 32 254 IE >=6.03. P7 7310-C04 for P5.x 2005.1 2006.06. P6.0 2010. P7 7042-C06 for P5. P6.12 R3.com/webapp/set2/sas/f/hmcl/resources. P7 7310-C03 for P5 & P6 7310-CR2 for P5 & P6 7315-C04 for P4 & P5 7315-CR3 for P4 & P5 7315-C03 for P4 & P5 7315-CR2 for P4 & P5 7315-C02 for P4 & P5 7315-C01 for P4 & P5 6878 V3 POWER7 (P7) POWER6 (P6) POWER5 (P5) POWER4 (P4) CFM Support – Note 21 Browser based GUI N Y N N Y N Y N Y N Y N Y N Y N Y N Y N N N N N N Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y Y N N Y Y Y Y Y Y N N N N N N Y Y Y .26 48 32 254 V7R3.html http://www14.14 R1. P6 7310-CR3 for P5.0 or IE >=6. P6.04. -CR2.02. HMC used to manage any POWER7 processor based server. P6.com/webapp/set2/sas/f/hmcl/home. © Copyright IBM Corp.com/webapp/set2/sas/f/hmc/home. -C04.0 or FF>= 1. -C07.html http://www14.x 2006.x 2007.06.software.com/webapp/set2/sas/f/power5cm/power7.x 2005.ibm.02. 2007 Page 17 of 40 AIX Facts and Features .0 2005.22 R1.

System i 570 (POWER5 processor-based) An HMC is optional for all other Power Systems servers. Utility CoD Multiple POWER5. and ZZZ is the last disruptive service pack level. processor-based servers can be managed by one HMC. including permanent activations.software. an installation is concurrent if: _ The service pack level (FFF) of the new firmware is higher than the service pack level currently installed on the system and the above conditions for disruptive installation are not met. the HMC machine code level must be updated to the server with the most recent firmware level. • http://www14. System p5 590. however. © Copyright IBM Corp. the HMC should have 3 GB of RAM minimum and be CR3 model. System i® 570 (POWER6™ processor-based). The HMC is used to manage partitions and support other critical functions on these servers: System p5® 570. System i 525. All prior levels of server firmware are supported with the latest HMC machine code level. deskside. and change partitions or CoD resources. System i 570 An HMC is recommended for managing all CoD. or later. An Ethernet hub will provide a physical connection from the HMC to each server rather than a logical Ethernet connection over a LAN. ES – Low End. Power 780 An HMC is required to manage partitions on the following servers and is recommended for other critical functions. Power 770. rack-mount or C06 model.1. 02 – Power Subsystem NN – Platform and class: EL. The following servers require an HMC. or Linux partition. directly controlling its I/O partition on the following servers: System i 515. EB or EP – Bulk Power SSS – Release indicator. Power® 560. define. IVM is part of the optional PowerVM™ virtualization software. the HMC must be a CR3 model. or later. Power 575. On/Off CoD. Customers choosing the HMC must use it to create. simple configurations without partitions or temporary CoD can be managed without an HMC: System p5 570. or later. Proxy-HTTP support for call-home is provided in 01SF240-258 (GA7SP3) and HMC V6. Firmware Levels • POWER7: 01ALXXX_YYY_ZZZ or 01AMXXX_YYY_ZZZ POWER6: 01EMXXX_YYY_ZZZ POWER5: 01SFXXX_YYY_ZZZ XXX is the release level. rack-mount HMC or C05 model. but it is not required to run the partitions. Otherwise. or later. EA. 2007 Page 18 of 40 AIX Facts and Features . DDD – Release or Service Pack level of the last Disruptive level An installation is disruptive if: _ The release levels (SSS) of currently installed and new firmware are different.ibm. including the following: Trial CoD. System p5 595. or later. EH – High End. either physically attached to the server or logically attached over a LAN.• Higher availability configurations may require the use of an Ethernet hub. System i 595.html Firmware Naming: PPNNSSS_FFF_DDD PP – Package Identifier: 01 – Managed System. Power 595. YYY is the service pack level. System i 550. If an HMC is used to manage any POWER7™ processor-based server. • • • • • • • • • • If a single HMC is attached to multiple servers. customers must choose either an HMC or IVM to manage partitions. An HMC is required for temporary CoD. For these systems.com/webapp/set2/sas/f/power5cm/home. Power 570. FFF – Service Fix pack. EM – Mid Range. System p5 575. deskside HMC. An HMC is required to manage more than one IBM i. _ The service pack level (FFF) and the last disruptive service pack level (DDD) are equal in new firmware. AIX. When IBM Systems Director is used to manage an HMC or if the HMC manages more than 254 partitions.

4500 tcp tcp tcp tcp tcp tcp tcp tcp tcp tcp udp/tcp udp/tcp tcp udp tcp tcp tcp tcp tcp tcp tcp tcp udp udp udp tcp tcp icmp udp WebSM WebSM ssh ssh rexec rexec Inventory Scout http https RMC RMC FCS FCS Web server Web server Secure Web Server Vtty I5250 console X11 CIM CIM Indication NTP l2tp SLP NFS TFTP Ping IPSec no no SSL 3DES Listen only Pair of ports used HMC Admin WS HMC Admin WS HMC LPARs yes yes yes yes yes yes SSL DLPAR LPARs ??? Call Home Call Home InfoCenter (V4) InfoCenter (V5) Remote ASM(V5) yes yes yes yes yes yes yes yes yes yes no yes no no no yes no yes yes yes yes yes yes yes no yes yes yes yes yes no no yes no SSL 2301.701 427 2. 03N5020 80P4543.03N7069 03N4698. 80P4383 09P4038. 03N5029. 00P4494. 2007 Page 19 of 40 AIX Facts and Features .411 4.443 9. 80P6455.920 9. NPIV PN 10N9824: CCIN 577D PCIe LC.SSL Xhost SSL SSL CIMOM provides inf.988 9197. 03N6439. 03N7067 00P4295. 09P0102. Converter cable LC-SC 50 Micron (#2456) or LC-SC 62. 9198 123 1. 80P3388. 03N5005 10N8620. Converter cable LC-SC 50 Micron (#2456) or LC-SC 62. 00P2995. to CIM clients Used in Cluster VPN HBA/FC Adapter Features FC 5708 8725 5735 5773 5774 5758 1905 5759 1910 5716 1957 1977 6239 6228 6227 Gb/Sec 10 10 8 4 4 4 4 4 4 2 2 2 2 2 1 Ports 2 2 2 1 2 1 1 2 2 1 1 1 1 1 1 Part # QLogic CNA/FCOE SFP+ Multimode Fiber SR optical or SFP+ Active Copper CCIN 2B3B CFFh for Blades QLogic 10Gb SFP+ SR Optical Transceiver. 09P1162. 80P4381.5 Micron (#2459) 03N5014. 24L0023 © Copyright IBM Corp.000 5. 09P5079.03N6440.900 4. 03N4167.03N6441.03N7068 80P6101.Power 5 HMC port information Port Protocol Application Enabled by Default Modification allowed in Network Config yes yes Security Notes 9090. 30000 –to.412 9.30009 22 1024-65535 512 & 514 1024-65535 1024-65535 808 80 443 657 697 1697 9. 80P6415 03N2452. 09P5080.049 69 n/a 500. 9940.735 2300. FC 0064 included with adapter PCIe Emulex LPe12002.5 Micron (#2459) PCIe LC. 2301 6. Part# 49Y4218.

SLP – Service Location Protocol.16 KB block size. of PPs in a PV instead the limit is at the VG level. J2-JFS2. one or more filesystems with a total size of 1 TB mounted. DGD-Dead Gateway Detection. CAPP/EAL4+: Controlled Access Protection Profile and Evaluation Assurance Level 4+ (CAPP/EAL4+) 3.1.0. one or more filesystems with a total size of 10 TB mounted .256 KB or greater block size. Max value for thewall : AIX5. Tested max.3 or later. Scalable VG removes the limit on the no. HMT-Hardware Multi Threading. GPFS filesystem size: depends on the block size of the filesystem . In AIX 53.1. WLM-Work Load Manager. SOD – Statement of Direction. . . Alternatives are 32 disks with 1016 PPs.64 KB block size. one or more filesystems with a total size of not greater than 200 TB where no single filesystem exceeds 100 TB mounted. Opt – Optional.1 thru v3.1. SMT–Simultaneous Multi Threading. Max PPs/disk: It is a combination of PVs in a VG. © Copyright IBM Corp. smaller of ½ of RAM or 1GB for 32b kernel.3 all service levels 32.2 file system architectural limit 2^51 bytes (2 Petabytes) Current tested limit approximately 2 PB.Notes 1. 16 disks with 1016*2 PPs. 6. of filesystems: GPFS v3.5 or later 64. 1 disk with 1016*32 PPs.0. 5. VIPA-Virtual IP Address.4 32 GPFS v2. .GPFS 2. 1GB or half the memory for CHRP. 7.1 and later – smaller of ½ of RAM or 64GB for 64b kernel.0. 2.No. GPFS 2. SUMA – System Update Management Assistant for policy based automatic download of updates. 256MB or ½ the memory for non CHRP. J-JFS. SCTP – Stream Control Transmission Protocol 4. GPFS v3. file system architectural limit 2^99 bytes. 2007 Page 20 of 40 AIX Facts and Features .

32 Licensed and 32 COD. is 1 x ViPr upto a max of 64. 20 VLAN/VE trunk adapter. of LiPr allocated to a shared processor pool. For file systems created prior to GPFS version 2. Ratio of 10 to 1 with the ShPr. LoPr-Off is 8. Ratio of 2 to 1 with ViPr. 17. LoPr-On 2xViPr upto a max of 128. of processors and memory configured. For file systems created prior to GPFS V2. of 64. the limit is 2. For file systems created with GPFS V2.147. 256 Virtual Ethernet connects/LPAR. if 8 Processors (DePr) in a processors allocated.456.3. 19.5 support is withdrawn effective Apr 30. 8. 32 Licensed and 32 COD. of Processors Licensed. 11. and the maximum disk size is only limited by the device driver support. Bonus Pack withdrawn as of May 21. 32 Licensed and 32 COD. . 16 VE trunk adapter/SEA. No. of Processors Licensed. DePr ShPr Page 21 of 40 Logical Processors SMT on (LoPr-On) in a LPAR Logical Processors SMT off (LoPr-Off) in a LPAR Entitled Capacity © Copyright IBM Corp. LoPr-Off x ViPr upto a max of 64. Let ViPr be 8. max 64. LiPr is 32. Please note that the effective limit on the number of files in a file system is usually lower than the architectural limit. PhPr is 64. of Processors configured in the server including COD option. ShPr is 8.3. 2008.No. from 1 to 8. 2007 AIX Facts and Features .com/servers/aix/expansionpack/ 10. LoPr-On is 16. Example: Out of 32 LiPr. http://www. of SEA/VIO – unlimited. LiPr+COD processors in the server. is 2xViPr upto a max of 128. If system firmware and HMC code is pre 10/2002. Example: If a server is configured with a total of 64 processors. IBM’s intent is to offer this feature in future and a statement of direction only. Example: If ShPr is 8. PhPr is 64. 13. Example: A server configured with a total of 64 processors. 320 VLAN/physical ENT. if 8 processors allocated. Dedicated Partition Total no. ViPr can be from 1 to 64. Example: DePr is 8. No.483. 2004. Maximum limit depends on the pSeries server Type-Model. then minimum physical memory for each partition is 1 GB. refer to the announcement letters or HMC code information URL for specifics.. Example: Out of 32 LiPr. LoPr-On is 16. products that previously existed on the AIX Bonus Pack now resides on the AIX Expansion Pack or the Web Download Pack.3 or later. On other supported platforms.Virtual Ethernet LAN. Example: ViPr is 8. 12.3 or later. VELAN/VEL . Example: ViPr is 8. of LiPr allocated to a dedicated a pool or Dedicated LPAR. the maximum disk size is 1 TB due to disk format limitations. max 64. GPFS supports disks up to 2 TB in size. Micro-Partition Total no. ViPr can be Example: If DePr is 8. LoPr-On is Ratio of 2 to 1 with ViPr. can range from 1 to ShPr upto a max. can LPAR range from one to DePr. Licensed Processors (LiPr) in No. of files in a filesystem: The architectural limit of the number of files in a file system is determined by the file system format. these limitations have been removed. Example: ViPr is 8. DePr is 8. max 64. No. 18. Example: A server configured with a the server or CEC. of Processors configured in the server including COD option. PSSP 3. dedicated LPAR Virtual Processors (ViPr) in a Ratio of 1 to 1 with the DePr. up to the operating system limit. max 64. pl. total of 64 processors.Disk size: The maximum disk size supported by GPFS depends on file system format and the underlying device support. Physical. Ratio of 1 to 1 with ViPr. Example: If a server is configured with a total of 64 processors.435. Logical and Virtual Processors: Processor Terminology Physical Processors (PhPr) in the server or CEC. LoPr-Off is 1 Ratio of 1 to 1 with ViPr. LoPr-Off is 8. Let ViPr be 8.648. For file systems created with GPFS 2. Designed to be compliant. 9. LiPr+COD processors in the server. GPFS supports disks larger than 2 TB (provided the disk device supports it). and could be adjusted using the -F option of the mmchfs command. no. SEA-Shared Ethernet Adapter in a VIO. LiPr is 32. the limit is 268. AIX 5L with 64-bit kernel.ibm. Maximum limit depends on the Power Systems/System p/pSeries server Type-Model. Shared Processors (ShPr) in No. 32 Licensed and 32 COD.

For more detailed information. the root user can create multiple instances of AIX on a single root volume group (rootvg). and customize this new instance of the Base Operating System (BOS). A new utility. called the standby BOS. 4x-10 Gb/sec. /usr/sbin/multibos. can be in production while multibos operations are used to modify the non-running instance.3. The running instance. Both of these device drivers support only 64-bit kernel mode. Filesystem Block Size 512 1024 2048 4096 Max. 12x-30Gb/sec.5 Gb/sec. is supplied in AIX 5L with 5300-03 to create and manage a new instance of the operating system within the running rootvg. You must run multibos with maintenance level 5300-03. of processors in the ViPr LPAR (No. of procs shown in lscfg or lsdev commands) 20.3 Commands Reference incorrectly lists the supported level for multibos as 5300-02. 22. and documentation regarding multibos in the AIX Information Center. maintain. P5-570 and 59x also require power subsystem code at 02BP230_125 prior to upgrading LIC. GX-4x & 12x with Virtualization across 64 LPARs. the multibos man page.multibos file.No. There are two InfiniBand device drivers: one for the GX bus and the one for the PCIX bus. Concurrent mode diagnostic support for the PCIX adapter is not provided. JFS2 filesystem size depends on the block size. The multibos utility provides the root user operations to setup. 2007 Page 22 of 40 AIX Facts and Features .0 and FW 01SF230_120_120 are the minimum code levels required to support CFM and IO Drawer concurrent maintenance. update. HMC V4 R5.PCI-4x only. 23. refer to the latest /usr/lpp/bos/README. 24. called the active BOS. The multibos command and corresponding man page in the AIX 5L Version 5. QDR-120Gb/sec IBM IB HCAs . Filesystem Size in TB 4 8 16 32 ViPr 21. The result of creating a new instance of the BOS with multibos is a rootvg with two distinct and bootable instances of the operating system within a single rootvg. access. Multiple Instances of AIX on a Single Root Volume Group In AIX 5. IVM – Integrated Virtualization Manager © Copyright IBM Corp. DDR-60Gb/sec. IB HCA Speeds & Feeds 1x-2.

2007 Page 23 of 40 AIX Facts and Features .1 Performance Management): AIX supports up to four different page sizes. No redundant VIOS partition support: Because all physical I/O must be owned by a single partition.ibm.software. The following table lists the page sizes supported by AIX and required System p™ hardware: Configuring the number of large pages: Use the vmo command to configure the number of 16 MB large pages © Copyright IBM Corp. Supported Memory page sizes (Ref AIX 6. The partition should be powered off first. it is not possible to have more than one VIOS partition for redundancy.html 26. Limited service function: There is no call-home support. all client partitions will also fail as their virtual disk.IVM may be used to complete the following tasks: Create and manage logical partitions Configure the virtual Ethernet networks Manage storage in the Virtual I/O Server Create and manage user accounts Create and manage serviceable events through Service Focal Point Download and install updates to device microcode and to Virtual I/O Server software Back up and restore logical partition configuration information View application logs and the device inventory Restrictions and Limitations Because the Integrated Virtualization Manager provides a subset of the HMC functionality. Full dynamic LPAR support for VIOS partition only: Dynamically adding or removing memory or processing resources from a running client partition is not supported with the IVM. All physical I/O is owned by the VIOS partition: This statement means that the VIOS partition is a single point of failure . and concurrent maintenance support is limited to adapters owned by the VIOS.software. but the actual page sizes supported by a particular system will vary based on processor type.ibm. there are some key restrictions and limitations that are worth highlighting. and Ethernet devices will not be accessible. optical. 25. lessening the importance of processing dynamic LPAR in most environments.if it fails. CSM limitations and FAQ http://www14. Keep in mind that the POWER5 Hypervisor allows partitions to use more than their entitled processing capacity via the shared processing pool.com/webapp/set2/sas/f/csm/home.com/webapp/set2/sas/f/csm/documentation http://www14.

and 64 KB pages for its stack on supported hardware: $ LDR_CNTRL=DATAPSIZE=64K@TEXTPSIZE=4K@STACKPSIZE=64K mpsize. the maximum size of the boot image has changed from the previous value of 11. 28. and text memory regions of a running process. The maximum real memory supported by a 32-bit kernel system (or partition) is 96 GB. the large page configuration changes will not take effect until you run the bosboot command and reboot the system. the following command causes mpsize. The vmstat command is enhanced to display information about multiple page sizes. You can specify the page sizes to use for three regions of a process's address space using an environment variable or settings in an application’s XCOFF binary with the ldedit or ld commands as shown in the following table: For example. The AIX 5L Version 5. System p. Assign 16 GB huge pages to a partition by changing the partition’s profile. or OEM hardware based on the Common Hardware Reference Platform (CHRP) architecture. You can use the pagesize -af command to display all of the virtual memory page sizes supported by AIX on a system. the following vmo command can be used to disable all kernel support for 64 KB and 16 GB pages: vmo -r -o vmm_mpsize_support=0 27. stack. The specific page sizes supported on a system depends on a system's processor type. 2. Furthermore.984 KB (32 MB minus 16 KB).984 KB (12 MB minus 16 KB) to 31. On systems that support dynamic logical partitioning (DLPAR). Maximum size of boot image increased: For AIX 5L Version 5. and no system configuration changes are necessary to enable a system to use these page sizes.2 is the last release of AIX that supports the uniprocessor 32-bit kernel. Kernel Support: The AIX 5L operating system previously contained both a uniprocessor 32-bit kernel and a 32-bit multiprocessor kernel. In addition to these three memory regions of a process's address space. The -p and -P options to the vmstat command displays VMM statistics for each supported page size. the support for 16 MB large pages and 16 GB huge pages is limited.3 32-bit multiprocessor kernel supports the following systems: RS/6000. and a system administrator must configure a system to use these page sizes. a process will continue to use 4 KB pages for all three process memory regions by default. you can select the page size for system V shared memory regions by using the SHM_PAGESIZE command to the shmctl() system call.on a system. the -r option can be omitted from the above command to dynamically configure 16 MB large pages without rebooting the system. The 4 KB and 64 KB page sizes are intended to be generalpurpose. The following example allocates 1 GB of 16 MB large pages: # vmo -r -o lgpg_regions=64 -o lgpg_size=16777216 In the above example. go to Properties → Memory → Advanced Options → Show Details to change the number of 16 GB pages. 1. The 16 MB large page size and 16 GB huge page size are intended only to be used in very high performance environments.3.3. regardless of the number of processors. © Copyright IBM Corp. 4 KB pages for its text.out Unless page sizes are selected using one of the above mechanisms. Finally. Configuring the number of huge pages: Huge pages must be configured through a system’s Hardware Management Console (HMC). On the managed system.out to use 64 KB pages for its data. The ps -Z command displays the page sizes being used for the data. 16 MB large pages are only supported for process data and shared memory. and 16 GB huge pages are only supported for shared memory. Effective with AIX 5L Version 5. 2007 Page 24 of 40 AIX Facts and Features . AIX 5L Version 5. the operating system supports only the multiprocessor kernel.

FSCNTL_REFREEZE. regardless of which kernel is being used. Reboot the system. 0). • Binary Compatibility. In addition. Dual-mode Kernel Extensions. Application Support. This feature provides an external interface whereby an application can request that a JFS2 file system freeze.The AIX 5L Version 5.3 supports dual-mode kernel extensions. rmts. and the path name of the multiprocessor versions of the 32-bit kernel is /usr/lib/boot/unix_mp. AIX 5L Version 5. The request for freeze or thaw can be performed from the command or from the API as follows: Command: chfs -a freeze=<timeout or "off"> <file system name> . lscosi. In addition. NULL. the 64-bit kernel is enabled during base system installation. A dual-mode kernel extension is an archive file that contains both the 64-bit and 32-bit versions of the kernel extension as members. Differences between 32-bit and 64-bit kernel systems are limited to the following: System and I/O Support. FSCNTL_THAW. swts.3 with the 5300-01 Recommended Maintenance package. (caddr_t)timeout. you can override this default option at installation time through the system installation panels. the 32-bit kernel is enabled. The 32-bit and 64-bit kernel systems have common base libraries. Binary compatibility is provided for 32-bit applications running on earlier versions of AIX on POWER-based systems. fscntl(vfs. Run the bosboot command to write a new system boot image. Installation and Enablement. On all other systems. Common Boot Image Management _ New commands and new SMIT interfaces – mkts. particularly the device drivers for the I/O. The 64-bit kernel limits support to 64-bit POWER-based systems. dbts. 29. JFS2 file system freeze and thaw feature: A new feature for the JFS2 file system is added to AIX 5L Version 5. 2007 Page 25 of 40 AIX Facts and Features . 1. 3. FSCNTL_FREEZE.3 64-bit and 32-bit kernel systems. commands. The 32-bit and 64-bit kernels are provided as part of the AIX 5L Version 5. so existing 32-bit kernel extensions may have to be ported in order to be used with the 64-bit kernel. After the freeze operation. 30. Some kernel services available in the 32-bit kernel are no longer provided by the 64-bit kernel. fscntl(vfs. However. cpcosi. except for applications linked statically or applications dependent on undocumented or unsupported interfaces.v Kernel Extensions. chcosi. 0). (caddr_t)timeout. existing 32-bit kernel extensions and device drivers used by 64-bit applications might have to be modified in order to support the new 64-bit ABI. in the absence of any application dependencies on internal kernel details or on kernel extensions that are not supported under the 64-bit kernel but are supported under the 32-bit kernel. The path name of the 64-bit kernel is /usr/lib/boot/unix_64. some system file formats have changed. but these kernel extensions are not usable by the 64-bit kernel. the 64-bit kernel does not support all I/O that is supported by the 32-bit kernel. You can switch between the 32-bit and 64-bit kernels without reinstalling the operating system. The 64-bit kernel supports both 32-bit and 64-bit applications. The default kernel enabled during installation is dependent on the hardware system being installed. On POWER5 systems. which can be loaded by a common configuration method. mkcosi. chfs -a refreeze=<timeout> <file system name> API: fscntl() fscntl(vfs. all 64-bit applications and libraries must be recompiled on AIX 5L Version 5. Application source and binaries are portable between AIX 5L Version 5. 0). In addition. and header files. and rmcosi © Copyright IBM Corp. or stay quiescent. the file system must remain quiescent until it is thawed or until the specified timeout has past. while the 32bit kernel supports both 32-bit and 64-bit POWER-based systems. AIX 5L Version 5. To take advantage of the scalability improvements to 64-bit programs. lsts. • Application Scalability. 2. regardless of which kernel is being used. utilities. Kernel extensions for the 64-bit kernel run in 64-bit mode and have the scalability of the larger kernel address space.3 base media and are installed on all supported hardware systems. Modify the /usr/lib/boot/unix directory and the /unix directory to be a symbolic link to the binary for the desired kernel.3 provides a more scalable application binary interface (ABI) for 64-bit applications.3.3 kernels provide the same functionality. and 32-bit applications processing these files might need to be recompiled. Existing 32-bit kernel extensions continue to be supported by the 32-bit kernel. Not all of the kernel extensions supported for the 32-bit kernel are supported for the 64-bit kernel.

printers. With a large number of devices in the system. Memory Segments • 32 bit address space: 16x256 MB for 32 bit. logical volumes._ New functionality – Clone a common image and modify the clone image. On the other hand. • AIX includes both logical devices and physical devices in the ODM device-configuration database. It can also require more device special files. Security Certifications: > Labeled Security Protection Profile (LSPP) to meet the assurance requirements of Evaluation Assurance Level 4 augmented (EAL4+). – Backup user data in /home. – Allow a thin server to switch to a different common image – Specify a time to switch to different common OS image. 33. assume that each path takes up the space of one-fifth of a device. The number of special files or i-nodes used also varies from device to device. Depending on the various factors mentioned in this section. and certain files in /root and restore this data after switching new common image.com/servers/aix/products/aixos/certifications/ 32. adapters. and enclosures. buses. Ref: AIX 6. These numbers include both physical and logical devices. • More devices require more real memory. SC23-6605.2 and the minimum RAM file system size. For example. only half as many disks could be attached. if you define a volume group with two logical volumes for each physical disk that is attached to a system. if you define a volume group with six logical volumes for each physical disk. The number of devices that AIX can support can vary from system to system. Logical devices include volume groups. In some cases.1 Operating system and device management. http://www-03. an MPIO device with five paths will have the space equivalent to two non-MPIO devices. • Some devices require more space than others in the ODM device-configuration database. • Multipath I/O (MPIO) devices require more space than non-MPIO devices because information is stored in the ODM for the device itself as well as for each path to the device. For example. With the maximum RAM file system size. > Controlled Access Protection Profile (CAPP) under the Common Criteria for Information Security Evaluation (CC) at the Evaluation Assurance Level 4+ level (EAL4+). – Allow thin servers to run a debug boot – Reduced user interface complexity 31.75GB Seg 0 : Kernel text & data Seg 3 to C & E : Shared memory segments for user processes Page 26 of 40 © Copyright IBM Corp. depending on several important factors. it is likely that up to 5000 AIX devices can be configured. the longer configuration time contributes to a longer boot time. Device drivers are located in the /usr/lib/drivers directory. it is likely that up to 25.ibm. the amount of space and i-nodes required of the file system depends on the types of devices on the system. The following factors have an impact on the file systems that support the devices: • Configuring a large number of devices requires storage of more information in the ODM deviceconfiguration database. max shared memory segments is 11 or 2. network interfaces. As a result. Therefore. your system might be able to configure more or fewer devices than this number. more space and more i-nodes are required of the file system. As a rough guideline. this will result in four AIX devices for each disk. With AIX 5. Page 262 Devices include hardware components such as. there will be eight AIX devices for each disk. the relationship between logical and physical devices can greatly affect the total number of devices supported. • Changing device attributes from their default settings results in a larger ODM device-configuration database and could lead to fewer devices that can be supported. /tmp. as well as pseudo-devices. and so on. 2007 AIX Facts and Features . As a result.000 AIX devices could be configured. such as the error special file and null special file. drives.

The encrypted data can be backed up in encrypted format. • Trusted Execution (TE): The Trusted Execution feature provides for an advanced mechanism for checking and maintaining system integrity. 34.Seg 1 : User Process text Seg 2 : User data. Such nonroot users will be able to assume the role and perform the allowed privileged operations.Default shared memory segments. © Copyright IBM Corp. reducing the risk of data being compromised if backup media is lost or stolen.448 PB) -Application text. The IT industry often uses COBIT IT Governance (Control Objectives for Information and related Technology.576 PB) .Privately loaded objects. The JFS2 encrypting filesystem can also prevent the compromise of data even to root-level users. AIX V6 comes with one predefined policy for assisting in SOX-COBIT compliancy setups and checks. data. Seg D : Shared library text Seg F : Per process shared library data • 64 address space (0 . which is a set of best practices for IT management created by the Information Systems Audit and Control Association (ISACA) and the IT Governance Institute (ITGI) in 1992) as the de facto standard to comply with the Internal Controls Section 404 of the Sarbanes-Oxley1 (SOX) act of 2002 (a United States federal law passed in response to a number of major corporate and accounting scandals).960 PB) . (960 PB . This enables any user on the system to exchange files in a secure manner if their counterpart offers this extension as well.Shared text and data (640 . Also administrators can define policies such that the loads of files listed in the database are monitored and execution/loads not allowed if hashes do not match.512 PB) . and heap. TLS is a cryptographic protocol that provides secure communication between clients and servers. These authorizations can be used to create the required roles necessary and assign those roles to the users required to manage the system. (576 . The TE tool can be used to check the integrity of the system against the database. 2007 Page 27 of 40 AIX Facts and Features . based on OpenSSL. including root. or low. • Secure FTP: AIX V6 introduces a secure flavor of ftp (and ftpd).1 Features • Role Based Access Control (RBAC): Role Based Access Control (RBAC) is designed to improve security and manageability by allowing administrators to delegate system administrative duties to nonroot users. Trusted AIX is the feature that provides the framework for Mandatory Access Controls and Labeled Security.System reserved. administrators can lock the signature database or the files in the database from being modified by any one in the system. Clients can select from a number of different encryption algorithms. RBAC in AIX has been enhanced to provide very fine granular authorizations that by name identify the privileged operation which they control. The command can be run at different levels with a flag to indicate the security levels of high.1 EB) . (512 .4 GB) . • Enhancements to the AIX Security Expert: The AIX Security Expert was introduced with Technology Level 5 update to the AIX V5. Trusted AIX is implemented as an installation option that can provide the highest levels of label-based security to meet critical government and private industry security requirements.3 operating system. A signature (SHA256/RSA) database for the important system files is created automatically as part of regular AIX install. • Encrypted File Systems (EFS): The IBM Journaled Filesystem Extended (JFS2) provides for even greater data security with the addition of a new capability to encrypt the data in a filesystem. AIX 6. medium.User process stack (Ref: Developing and Porting C and C++ Applications on AIX and General Programming Concepts: Writing and Debugging Programs). (448 . heap & stack. The roots of Mandatory Access Control (MAC) systems come from the Department of Defense Orange book standards developed in 1985. using Transport Layer Security (TLS) (This extension to FTP is defined in RFC 4217) to encrypt both the command and the data channel. (4 GB . AIX 6 includes an enhancement to the Security Expert to store security templates in a Lightweight Directory Protocol (LDAP) directory for use across a client's enterprise. • File Permission Manager (FPM): fpm can reduce the security risk of the set UID permissions by removing these permissions. Additionally.640 PB) . • Trusted AIX: Trusted AIX extends the security capabilities of the AIX operating system by supplying integrated multi-level security. and provides clients with the capability to manage more than 300 system security settings from a single interface and the ability to export and import those security settings between systems.First 16 segments are exempt from general use to keep compatibility with 32-bit user process model.

It can be used to see some of the system state and do some debug in the case that a full dump is not available.• Secure by Default (SbD): The AIX 6 installation process will offer a new option.BZ. enterprise RAS defines a component framework where AIX and third party software can register components that enable their specific RAS features such as trace. – Dynamic tracing with probevue: AIX 6 will provide a new dynamic tracing capability that can simplify debugging complex system or application code without requiring code changes and recompilation. when a system dump happens on a multiprocessor system. The Component Trace facility (CT). that allows a developer or system administrator to dynamically insert trace breakpoints in existing code without having to recompile the code.1 operating systems. Moreover. because selective dump aware components can be chosen. This increases reliability of a partition system dump by minimizing the work done by the failing operating system and let it © Copyright IBM Corp. Thus. This dynamic tracing facility will be introduced via a new tracing command. the blocks are compressed and unordered: this unordering feature allows multiple processors to dump in parallel sub-areas of the system. to be used with the -L option for the statistical information on the most recent dump is also added to sysdumpdev command. dump and error checking features. starting with AIX V6 The firmware-assisted dump means that AIX dump is taken while the partition is restarting. that installs only the minimal number of services to provide the maximum amount of security. starting with AIX V6 Live dumps are small dumps that do not require a system restart. This new capability will provide IBM with a tool to reduce the number of unplanned outages required to maintain a secure. registered as a live dump enabled component. It can also be used to get a quick snapshot of a crash without having to transfer the entire dump from the crashed system. The Secure by Default option works particularly well when used in conjunction with the AIX Security Expert to only enable the system services required for the system's intended purpose. For additional information on these facilities. A new live dump feature allows selected subsystems to dump their memory state and traces to the filesystem for subsequent service analysis. reliable system. AIX operating system. – Live dump: AIX 6 continues to build upon the first failure data capture and nondisruptive service aid features introduced in prior AIX releases. > Component Dump facility. A developer or system administrator can use probevue to dynamically place probes in existing code.3 or AIX V6. Storage keys can reduce the number of intermittent outages associated with undetected memory overlays inside the kernel. In order to provide a granular approach to RAS. and then written to the error log on reboot. > Live Dump Facility. – Kernel Storage Keys: Kernel exploitation of the POWER6™ processor storage key feature brings a mainframe-inspired reliability capability to the UNIX market for the first time. starting with AIX V5. This new file format for parallel dump is no more readable with usual uncompress and zcat commands: the new dmpuncompress command must be used. The -S option scans a specific dump device and see if it contains a valid compressed dump. In this new compressed dump file. The dump file extension for this new format is still . > Minidump facility. Software or system administrators can initiate Live Dumps while the system is running: planned downtime is no longer necessary to dump a system. named a Live dump aware component.3 TL05 A new optimized compressed dump format is introduced in AIX V5. In order to increase dump reliability. The Component Dump acility (CD) for the AIX operating system components is now introduced with AIX Version 6. starting with AIX V5. and also for external third party software. see AIX 5L Differences Guide: Version 5. Applications can also use the POWER6 storage key feature to increase the reliability of large. The Live Dump facility uses the Component Dump implementation to dump only AIX components. like the Run-Time Error checking (RTE) facility have been implemented for the AIX Operating system components with AIX V5. probevue. • Continuous availability: AIX 6 includes many mainframe-inspired continuous availability features. > Parallel dump. 2007 Page 28 of 40 AIX Facts and Features .3 TL05. without requiring a full system dump and outage. > Firmware-Assisted Dump Facility. starting with AIX V6 The enterprise Reliability Availability Serviceability strategy is to maintain the continuous availability of System p servers through extended key error detection and recovery capabilities implementing mainframe-like features for the hardware. a new -S checking option. Secure by Default.3 TL03 The minidump is a small compressed dump that is stored to NVRAM when the system crashes or a dump is initiated. including: – Concurrent AIX updates: Concurrent AIX updates provides a new capability to deliver some kernel updates as Interim Fixes that will not require a system reboot to be put into effect. SG24-7414. and specify the data to be captured at probe point. Parallel dumps are produced automatically when supported by the AIX release. Live dump facility reduce significantly the time to dump and the size requirement for dump files.3 TL05.3 Addendum. complex applications running under the AIX V5. the time to produce the dump image is now I/O bound limited and so greatly reduced.

The following new tunables are introduced to change the behavior of the recovery manager. a new. > The live dump. graphical installation tool and a new graphical systems console. As shown in Figure 4-2 on page 106. © Copyright IBM Corp. based on the new dump components. including NFSv4 support for the Network Installation Manager (NIM). With AIX Version 6.done by the new restarting instance. Thus. the only supported type of dump was the system dump that requires a reboot afterwards. based on these components. * The raso command tunables With the raso command. to manage the attributes of these two different types of dumps. is called live dump. the Systems Director Console for AIX. this new dump capability. • Improving manageability: AIX 6 includes many new capabilities to improve the manageability of the AIX operating system. The firmware is involved to preserve memory area across the reboot. the implementation of the AIX dump components provides an enhanced dump granularity and allows to dump these components without requiring any reboot. you can manage reliability. is implemented with the component infrastructure that allows to take a dump while the server is running. 2007 Page 29 of 40 AIX Facts and Features . AIX provides a unified user-interface he dumpctrl command to manage both: > The traditional dump also named system dump that requires a reboot. availability. and serviceability parameters. The Systems Director Console for AIX is included with AIX 6. Before AIX Version 6. All recovery tunables are restricted and should not be changed unless requested by the IBM service support. The Systems Director Console for AIX provides a responsive Web access to common systems management ools such as the Systems Management Interface Tool (SMIT) and offers integration into the IBM Systems Director. The dumpctrl command is a new integrated interface to manage the various dump formats.

Refer AIX 6.0 introduces the 64-bit kernel (unix_64).1 Differences Guide for details. 2000) AIX V5. and nfso) have a unified behavior and syntax. they are no longer displayed by default but only with the -F option (force) on the command.* Restricted tuneables: Since AIX 5. schedo. Beginning with AIX Version 6. In previous AIX versions multiple kernels were shipped. © Copyright IBM Corp.3 and AIX V5. 2004) The uniprocessor kernel (unix_up) has been removed in AIX V5.1 kernel will provide 32 bit binary compatibility for applications running on AIX 5L on POWER4 systems and later. 2007 Page 30 of 40 AIX Facts and Features . The important milestones are: 1993) AIX V4. six tuning commands (vmo. 35: Only 64b Kernel in AIX 6. some tunables are now classified as restricted use tunables.1: AIX 6. no.3. ioo. They exist and must be modified primarily for specialized intervention by the development support or development teams. As these parameters are not recommended for end user modification. Thus.1 introduces multiprocessor kernel (unix_mp). In AIX V5. 32-bit kernel extensions and device drivers will not run and this does not affect support for user space 32 and 64 bit applications.2 ML 5200-03 the 64-bit kernel was installed by default on POWER5 and newer systems. in SMIT and Web-based System Manager they have no visibility by default. raso. The 32-bit kernel will no longer be available with AIX 6.3.

These features are enabled by default. the AIX operating system will simplify its kernel environment by providing only the 64-bit kernel.ibm.1 but only in 64-bit mode. and Run-Time Error Checking (RTEC).com/systems/p/os/aix/compatibility/index.1 Differences Guide): – Selected changes to AIX tuning defaults to optimize on common System p workloads – Primarily expected to benefit workloads sensitive to VMM tuning – Based on extensive benchmarking and tuning experience from AIX performance – New categorization of tunables into restricted and non-restricted types 37: Advanced First Failure Data Capture features AIX 5L Version 5. Ref – AIX 6. AIX V6.1 Beginning with AIX V6. Dual-mode (32/64-bit) kernel extensions built on AIX 5L will continue to run on AIX V6. Component Trace (CT®).1. These features include Lightweight Memory Trace (LMT).1 is binary compatible to both 32-bit and 64-bit applications created in previous AIX 5L versions. Further information and a list of restrictions are published on the IBM AIX Binary compatibility site: http://www-03. 2007 Page 31 of 40 AIX Facts and Features .3 with the 5300-05 Technology Level package provides many advanced First Failure Data Capture (FFDC) features.html 36: Solution Performance Tuning (out of the box performance. The advanced FFDC features can © Copyright IBM Corp. Device drivers and kernel extensions that are 32-bit only are not supported. at levels that provide valuable FFDC information with minimal performance impacts.2007) The 32-bit kernel is removed in AIX V6.

The impact on the throughput of a kernel-intensive benchmark is one percent. For the 64-bit kernel. All Run-Time Error Checking can be persistently disabled across reboots by running the errctrl P errcheckoff command. RTEC provides service personnel with powerful first failure data capture and second failure data capture error detection features. Component Trace The Component Trace (CT) facility provides system trace information for specific system components. For more © Copyright IBM Corp. 2007 Page 32 of 40 AIX Facts and Features . Recorded events are saved into per processor memory trace buffers. System trace The system trace facility has been enhanced to support process and thread-based tracing. then tracing runs continuously. CT can be persistently enabled by running the ctctrl -P memtraceon command. RTEC can be re-enabled persistently by running the errctrl -P errcheckon command. The use of in-memory CT buffers can be persistently disabled across reboots by using the ctctrl -P memtraceoff command. enter the following shell command: echo mtrc | kdb | grep mt_total_memory The raso tunable command can be used to disable LMT. Lightweight Memory Trace The Lightweight Memory Trace (LMT) provides system trace information for First Failure Data Capture (FFDC). see the trcctl command documentation. Note: A bosboot is required to make the command persistent on the next boot Information on these and other CT commands can be found in the ctctrl command documentation. one to record common events. Any change to LMT will not take effect until the next boot. The 32-bit kernel uses the same default buffer memory size calculation. This information allows service personnel to access component state information through either in-memory trace buffers or through traditional AIX system trace. It can also be used to increase or decrease the memory trace buffer sizes. enter the following command: smit ffdc You can then choose to enable or disable FFDC features. The following table shows some examples of default LMT memory consumption: To determine the amount of memory being used by LMT. The system trace can be used to trace processor utilization register (PURR) to provide more accurate event timings in a shared processor partition environment. Run-Time Error Checking The Run-Time Error Checking (RTEC) facility provides service personnel with a method to manipulate debug capabilities that are already built into product binaries. Note: A bosboot is required to make the command persistent on the next boot. To enable or disable all three advanced FFDC features. The memory trace buffers can be extracted from system dumps and accessed on a live system by service personnel. For more information. CT is enabled by default. The system activates LMT at initialization. The default amount of memory required for the trace buffers is calculated based on factors that influence software trace record retention. and is much less for typical user workloads. The 64-bit kernel resizes the LMT trace buffers in response to dynamic reconfiguration events. There are two memory trace buffers for each processor. The trace command supports settings of larger trace buffers for regular users. see the raso command documentation. You can restrict the tracing to a process and capture the events in relation to the process for better debugging. It is a constant kernel trace mechanism that records software events occurring during system life.be individually manipulated. For more information. but restricts the total memory allocated for LMT (all processors combined) to 16 MB. For more information. and no more than 256 MB by a single processor. the default calculation is additionally limited such that no more than 1/128th of system memory can be used by LMT. Note that a bosboot and reboot are required to fully enable or disable all FFDC features. LMT requires the consumption of a small amount of pinned kernel memory. the 32-bit kernel does not. see the trace command documentation. and one to record rare events.

Note 38: Restrictions using HEA under EtherChannel Host Ethernet Adapter logical ports are only supported under EtherChannel if all adapters within the EtherChannel are HEA logical ports (including the backup adapter. Xmalloc debug In AIX 5L Version 5. Note that each of these error log entries might identify a unique potential problem. Xmalloc debug is similar to the previous memory overlay detection system (MODS). Volume 2. see AIX 5L Version 5. having HEA logical port adapters intermixed with physical Ethernet adapters or Virtual I/O Ethernet Adapters in the same EtherChannel is not supported.5 seconds is logged per boot (default).3 Commands Reference. you must create an EtherChannel for the partition over both of the logical HEA ports. the physical ports associated with the HEA logical ports must also be placed in an EtherChannel in the Ethernet switch. all partitions that use HEA logical ports going to the same HEA physical ports must also be placed in an EtherChannel. For more information. © Copyright IBM Corp. this size error log entry is intended only for restricted system use and general error log entries should continue to contain 2048 bytes or less of event data. buffer overruns and accesses to freed data. To enable xmalloc debug. the following functions can be called from a disabled code section of a detected kernel extension to exempt the section from future excessive disablement reporting: disablement_checking_suspend disablement_checking_resume For more information about disablement checking. To specifically disable the xmalloc debug RTEC feature. RTEC features include: 1. AIX error logging now supports up to 4096 bytes of event data (see the /usr/include/sys/err_rec.3 with 5300-05 Technology Level. run the errctrl errcheckoff -c alloc. For more information.xmdbg -r command. Excessive Interrupt Disablement Detection The Excessive Interrupt Disablement Detection mechanism in AIX can detect whether or not privileged code remains disabled for interrupts for too long. Consequently.xmdbg -r command. Other RAS enhancements The chcore command provides for management of location of core files. The largest detail data size guaranteed not to be truncated is 2048 bytes. Also see the errctrl command documentation. For example.information on the errctrl command. Only one period of interrupt disablement that exceeds . this size entry may be truncated across a reboot in certain circumstances. Because excessive disablement might lead to performance problems. Attempting to use either of those logical HEA ports as stand-alone in other partitions might cause connectivity problems. These error reports are persistently disabled if RTEC is globally disabled.3 Technical Reference: Kernel and Subsystems Volume 1. When using multiple HEA logical ports as primary adapters in an EtherChannel. because packets might not be delivered to the correct logical HEA port. AIX writes an error log record to report this detection: IDENTIFIER TIMESTAMP T C RESOURCE_NAME DESCRIPTION A2205861 0705170705 P S SYSPROC Excessive interrupt disablement time Report these error logs to IBM service. see the MODS and errctrl command documentation. 2. random sampling of xmalloc allocations is enabled to catch memory leaks. 2007 Page 33 of 40 AIX Facts and Features .disa Finally. if any). similar to the configuration of Partition 1. Consequently. run the errctrl errcheckon -c alloc. disablement detection can be disabled by running the following command: errctrl errcheckoff -c proc. On a per-boot basis. While up to 4096 bytes of detail data is allowed. The detailed report contains additional information including a stack traceback and LMT (trace) data that can be used by IBM to identify the source of the problem. However. see ″disablement_checking_suspend Kernel Service″ and ″disablement_checking_resume Kernel Service″ in the AIX 5L Version 5. A large error log entry reduces the non-volatile storage available to the system dump facility in the event of a system crash. see the chcore command documentation. assume that Partition 1 is configured as follows: * A logical HEA port out of physical HEA port 0 * A logical HEA port out of physical HEA port 1 * An EtherChannel created using the logical HEA ports listed above If another partition on the same system that needs to use a logical HEA port out of physical HEA port 0 or out of physical HEA port 1.h file).

since the physical HEA ports do not require specific configuration on the Ethernet switch.The aforementioned configuration restriction does not exist for using logical HEA ports in a Network Interface Backup configuration (1 primary and 1 backup).1 IGMPv3 implementation adheres to RFC 3376 and includes the new Socket Interface Extensions for Multicast Source Filters. It allows hosts to participate in IP multicasting according to RFC 1112. Security Features and threats (Ref: Advanced Security SG24-7430) 39. A sender does not have to be a member of a multicast group to send packets to such a group. 2007 Page 34 of 40 AIX Facts and Features . The AIX V6. where the basics and specifics are described. AIX V6.1 provides the host side function and group member part of the IGMP Version 3 (IGMPv3) protocol.1 IGMPv3 implementation allows backward compatibility © Copyright IBM Corp. performing both the multicast router part and group member part of the protocol. 38. The AIX V6. IGMP: Internet Group Management Protocol (IGMP) is the protocol used by hosts and multicast routers to establish multicast group memberships within a physical network. The IGMP protocol allows routers to act as members of one or more multicast groups.

with the previous two versions of the protocol. /usr/bin/tsh. IGMP version 1 (IGMPv1) and IGMP version 2 (IGMPv2)./usr/bin/sh. as they are supported in AIX 5L releases./usr/bin/csh. Garbage collection on paging space blocks after a re-pagein The method of freeing a paging-space disk block after a page has been read back into memory from paging space is employed by default./usr/bin/ksh93.3./usr/bin/rksh93./bin/csh./usr.cfg shows that the administrator chose to use the "ssha256" LPA as the system wide password encryption algorithm: shells = /bin/sh. /bin/bsh./bin/ksh93./bin/ksh. /rksh./usr/sbin/uucp/uucico./usr/sbin/sliplogin. you can use the paging space garbage collection feature to free up paging-space disk blocks under certain conditions so that you do not have to configure as much paging space as the amount of virtual memory used for a particular workload./usr/bin./usr/bin/ksh./usr/sb in/snappd maxlogins = 32767 logintimeout = 60 maxroles = 8 auth_type = STD_AUTH pwd_algorithm = ssha256 Note 41: Paging space garbage collection: Starting with AIX 5./bin/bsh. 40./bin/tsh. 2007 Page 35 of 40 AIX Facts and Features . The reason that this is not freed up for every re-pagein is because leaving the blocks in paging space provides better © Copyright IBM Corp. Loadable Password Algorithm: The following example of /etc/security/login. The garbage collection feature is only available for the deferred page space allocation policy.

Very high speed adapters are set to 256 KB. rpgclean. A ″blank″ value means the option is not set so it would inherit the global ″no″ setting. 2007 Page 36 of 40 AIX Facts and Features . npsrpgax. rpgcontrol. it is not necessary to perform the re-pageout function.performance in the case of unmodified working storage pages that are stolen by the LRU daemon. You can tune the following parameters with the vmo command: npsrpgmin. Note 42: Network tuning: Use the following settings to enable rfc1323 if the MTU size is 4096 bytes or larger and to set the tcp_sendspace and tcp_recvspace values to at least 128 KB for high speed adapter (gigabit or faster). If pages are stolen. © Copyright IBM Corp.

Virtual I/O Server Version 2.4. Note 44: Beginning with AIX 5L Version 5.Enterprise Edition of PowerVM . Note 45: AIXTHREAD_SCOPE={P|S}. (AIX 6.The capability to reactivate a POWER6 595 processor book that has been repaired without powering down a Power 570 or 595.Firmware level of 340_070 (eFW 3. 2007 Page 37 of 40 AIX Facts and Features . Dynamic Adapter Membership functionality is available.The capability to: 1) deactivate.0 Service Pack 2 .HMC version V7R3. 2) repair components or add memory.1. • Hot-node Repair & Hot memory add .1 Differences Guide SG-24-7559) Note 46: Active Memory Sharing minimum requirements and limits (Ref. This functionality allows you to add or remove adapters from an EtherChannel without having to disrupt any user connections.1 for both HMC and IVM managed systems . then that node must be repaired via Cold-node Repair.The capability to add additional processor books to POWER6 processor-based 570 & 595 systems without powering down • Cold-node Repair . Hot-node repair (Ref Power 595 announcement and the White paper Highly Available IBM Power Systems Servers for Business-Critical Applications): • Hot-node Add .2 with the 5200-03 Recommended Maintenance package.Note 43: Hot-Node add.2) . * Minimum requirements: . P signifies process-wide contention scope (M:N) and S signifies system-wide contention scope (1:1). and then 3) reactivate a POWER6 595 processor book or POWER6 570 node1 without powering down.1-TL03 or SUSE Linux Enterprise Server 11 © Copyright IBM Corp. Either P or S should be specified and the current default is processwide scope. • If the POWER6 570 node being repaired is driving the system clocks. Redpaper PowerVM Virtualization Active Memory Sharing REDP-4470). Cold-node repair.4.An IBM Power System server based on the POWER6 processor .1 or AIX 6.OS: IBM i 6.

This limit has been in enhanced in PowerVM 2. © Copyright IBM Corp.3 – Announcement Apr 2010. This redundant paging VIOS configuration improves the availability of the VIOS to shared memory partitions.Shared Memory LPARs must have all resources virtualized (shared processors.In HMC-managed systems. Virtual Ethernet or SEA. * Processor and Memory Virtualization comparison.. Virtual Serial adapters) * Limits: . Virtual SCSI or VSCSI. Virtual HBA or Fibre Chaneel adapters. 2007 Page 38 of 40 AIX Facts and Features .The paging Virtual I/O Server is designed to handle up to 128 shared memory partitions.1. . up to two paging VIOS partitions can be assigned to a shared memory pool to provide access to the paging devices.

rte.wpars. modification.1/fqm0_r_os_supp orted_by_ibm_director_611. and fix level of the xlC.0 or later Firmware: 7. bos. Optional tests to check the compatibility of arrival and departure systems: • NTP must be enabled on both the systems.1/fqm0_r_support ed_hardware_and_software_requirements. the critical compatibility tests check the following compatibility criteria on the arrival system and the departure system: • The operating system type. 2007 Page 39 of 40 AIX Facts and Features .plan_6. depending on which WPAR group the WPAR belongs to and which optional tests were selected as part of the WPAR group policy.rte.ibm.html http://www-03. The critical tests for static relocation are a subset of the tests for live relocation. For each relocation type. release.html http://publib.1 TL4 SP2 or later © Copyright IBM Corp.rte filesets must be the same on both the systems. In addition to these critical tests additional optional tests can be added for determining compatibility. Two managed systems might be compatible for one WPAR and not for another.html 49.com/systems/management/director/downloads/ http://publib. the WPAR Manager starts a background process that compares the properties of the new system to the system properties of all previously registered systems.com/infocenter/director/v6r1x/index. Active Memory Expansion (AME) Hardware & software requirements: POWER7 server running in POWER7 mode HMC: V7R7.boulder. VMControl and Network Control are dependent on IBM Systems Director and have the same hardware and OS requirements as IBM Systems Director. • The version.rte file set must be the same on both the systems. Each time a system is registered.47.ibm.1 AIX 6.1.ibm. • Physical memory on the arrival system must be at least as high as memory of the departure system.jsp?topic=/director.libc file must be the same on both the systems.0.jsp http://www-03. release.com/infocenter/director/v6r1x/index. • The processor class on the arrival system must be at least as high as the processor class of the departure system. modification. VMControl and Network Control H/W and OS Support: IBM Systems Director Active Energy Manager. These optional tests are selected as part of the WPAR group policy for the WPAR planned for relocation. • The bos. For live relocation. there is a set of critical tests that must pass for one managed system to be considered compatible with another. Compatibility testing for application mobility (Ref: InfoCenter) Compatibility testing includes critical and optional tests. • The processor speed for the arrival system must be at least as high as the processor speed for the departure system. These requirements are available from the IBM Systems Director Information Center at http://publib.boulder.ibm. live or static.ibm.com/systems/management/director/plugins/ http://www-03. Active Energy Manager (AEM). • The version.boulder. version and level must be the same on both the systems. These compatibility tests help to determine if a WPAR can be relocated from one managed system to another. and are taken into account for both types of relocation. and fix level of the bos. This is the only critical test for static relocation • There must be at least as many storage keys on the arrival system as on the departure system.plan_6. mcr.jsp?topic=/director.com/infocenter/director/v6r1x/index. 48.com/systems/management/director/downloads/plugins/index. Critical tests are always applied in determining compatibility regardless of the WPAR group to which the WPAR belongs.ibm.

1 are supported only when used within the system operating environments described in the appropriate hardware announcements and when used within the specified operating environment. System p5®. compartmentalized security • Integrated Role Based Access Control • Encrypting JFS2 file system • Kernel exploitation of storage keys for greater reliability • Concurrent AIX kernel updates. Limitations AIX Express Edition is limited to a maximum of four processor cores in a partition with a maximum of 8 Gb of memory per processor core. and AIX Enterprise Edition in the same server.1 are used with other software or software in later announcements. memory. 2007 Page 40 of 40 AIX Facts and Features . When systems operating on AIX 6. Software requirements Systems operating on AIX 6. System p®. or POWER7 processors. System i®.50. Clients can mix partitions with AIX Express Edition. including the IBM Power Systems. eServer™ p5. which can reduce the need for planned outages • Dynamic tracing with probevue • Robust journaled file system and Logical Volume Manager (LVM) software. AIX 6 Express Edition includes all features of AIX 6 Standard Edition: • IBM PowerVM™ Workload Partitions Manager™ for AIX -. and I/O resources • Support for dedicated and shared processor LPAR groups • Trusted AIX: Multi-level. other limitations may be included. and eServer pSeries® server product lines. AIX Standard Edition. Clients can run multiple partitions with AIX Express Edition in a single server as long as each partition is running no more than four processor cores.System Management Interface Tool (SMIT) and the IBM Systems Director Console for AIX Hardware requirements IBM systems that run on the IBM POWER5™. © Copyright IBM Corp.software-based virtualization • Live Application Mobility with the IBM PowerVM Workload Partitions Manager for AIX (5765-G83) • 64-bit kernel for higher scalability and performance • Dynamic logical partitioning and Micro-Partitioning™ support: • Automated load balancing • Online addition or removal of processors. including integrated file system snapshot • Tools for managing your systems environment -. POWER6™.

Sign up to vote on this title
UsefulNot useful