Accelerate with IBM Storage: Spectrum Virtualize Copy Services

Transcription

1 IBM North America Technical Leadership Team : Spectrum Virtualize Copy Services Byron Grossnickle: Consulting I/T Specialist - Virtualization Bill Wiegand: Consulting I/T Specialist - Virtualization

2 Agenda Volume Mirroring FlashCopy Metro Mirror Global Mirror Global Mirror with Change Volumes 2

3 Volume Mirroring Volume mirroring allows creation of one Volume with two copies of MDisk extents There are not 2 volumes with the same data on them Copies can be in different storage pools In SVC that can mean across 2 storage frames Copies can be of completely different structures Image, striped, sequential, space-efficient, compressed Minimal impact to volume availability if one set of MDisks goes offline Automatic incremental resync Volume mirroring sits below the upper cache and copy services FlashCopy, Metro Mirror, Global Mirror have no awareness that a Volume is mirrored Anything that can be done today with a volume can be done with a mirrored volume including migration, expand/shrink, etc. In version 7.3 mirroring can take advantage of the lower cache Volume mirroring can be used for: Data Protection Data Transformation 3

4 Copy Rate On Initial Sync/Re-Sync Mirror Rate Setting Data Copied/s KB KB KB MB MB Default MB MB MB MB MB 4 9/1/2016

5 Spectrum Virtualize FLASHCOPY 5

6 IBM Storwize - FlashCopy FlashCopy Relationships Up to 256 targets FlashCopy: Volume level point-in-time copy with any mix of thin and fully-allocated Start incremental FlashCopy Data copied as normal Later Source vdisk Map 1 Map 2 Disk0 Disk1 Source FlashCopy target of Disk0 Disk3 FlashCopy target of Disk1 Map 4 Incremental FlashCopy: Disk2 FlashCopy target of Disk1 Volume level point-in-time copy Disk4 FlashCopy target of Disk3 Some data changed by apps Cascaded FlashCopy: Copy the copies Start incremental FlashCopy Only changed data copied by background copy

7 FlashCopy Overview Source volume may be spread across multiple disk subsystems or pools Target volume may be to one or more disk subsystems, different than source Create up to 256 copies of a source volume Because we use linked lists the source volume is only impacted by 1 copy on write operation regardless of the number of copies Copy is accessible almost immediately Copy may be updated independently of the source Background Copy Rate establishes copy rate goal Value of 0 has a "No Copy" effect Data rate 128 KB/s to 64 MB/s per Volume Support for 128 Consistency Groups Support for 4096 mappings Invoked via Web User Interface, CLI or script With 7.5 code 2 GB of FlashCopy bitmap space Enough to FlashCopy 4 PB using a 256 KB grain size Prior to 7.5 enough to copy 1 PB using a 256 KB grain size 7

8 Types of FlashCopy 8

9 Snapshot This is also known as a FlashCopy no copy Only stores changed grains from the original A grain is 64 or 256 kb in size Default is 256 kb 64 kb grain size uses 4X the bitmap space but can save physical disk space Background copy rate will be set to 0 Raising the background copy rate will produce a full copy Will generally always show copying in the status 9

10 Clone Produces a full, standalone copy of the original The target can be used for read/write immediately The Background Copy Rate determines the speed of the background copy Default of 50 (2 MB/s) Relationship is automatically deleted once the copy is finished 10

11 Backup (aka Incremental FlashCopy) Produces a full copy of the original The target can be used for read/write immediately Tracks changes from the original When started again, only transfers the changes to update the full copy to the new point in time of the original The Background Copy Rate determines the speed of the background copy Default of 50 (2 MB/s) Relationship is maintained once copy is finished 11

12 Automated FlashCopy by Right Click Automatically creates a target of the same type as the original Puts the copy in the same pool as original 12

13 Advanced FlashCopy Allows you to use existing target volumes OR create target volumes of the type you designate Allows you to designate the target pool ( if you choose to create them) 13

14 Background Copy Rate vs Clean Rate Background Copy Rate The rate at which data is copied to the target when the background copy rate is > 0 Default is 50 (or 2 MB/s) for non-snapshot copies and 0 for snapshots A FlashCopy nocopy can be changed to a full copy by raising the copy rate to something > 0 Clean Rate The rate at which data is copied between copies in a linked list when the clean rate is > 0 and the background copy rate = 0 Default is 0 for snapshots and 50 (or 2 MB/s) for non-snapshot copies If clean rate is > 0, each target gets grains that have been changed from the original to maintain it's point in time. This, in turn, means that in a multiple copy situation, each target may contain many of the same grains as other targets Eliminates stop time If clean rate = 0, each target only gets grains that are unique to it's copy that no other copy has. This saves space, but can extend the amount of time that it takes to stop a FlashCopy Saves space 14

15 Background/Clean Copy Rate Copy rate setting Data copied/s 256 KB grains/s 64 KB grains/s KB KB KB MB MB MB MB MB MB MB MB *By RPQ MB *By RPQ 15

16 FlashCopy Overview 1 Multiple targets are arranged in a linked list Most recently started mapping is at the head and dependent on source Targets started earlier are dependent on newer target mappings Only dependent until fully copied Write to source causes write to newest target in linked list Older targets get their data from newer target for unsplit grains Write to target causes write to next older target in linked list Assumes there is still a dependency on that target Example assumes clean rate of 0 on snapshots Use the lsfcmapdependantmaps command to figure out where dependencies exist 16

17 FlashCopy Overview 2 17

18 FlashCopy Overview 3 It is important to note that when using GMCV in combination with FlashCopy, the best practice is to stop GMCV, start your FC and then restart GMCV This ensures that the GMCV relationship is at the top of the dependency tree and will stop and start freely with the cycling mode. 18

19 Day1 - FlashCopy #1 w/clean rate = 50 19

20 Day 2 - FlashCopy Mapping #2 w/cleanrate = 50 After starting the FlashCopy #2 mapping change the block B to B 20

21 Day 3 - FlashCopy Mapping #3 w/cleanrate = 50 After starting the FlashCopy #3 mapping change the block C to C 21

22 Starting FlashCopy Mapping 1-3 w/cleanrate = 0 22

23 Single Platform for Remote Copy Services Single point for copy services Use to meet business needs Asynchronous remote copy/global Mirror Disaster recovery up to 25000KM Synchronous remote copy/metro Mirror Business Continuance up to 300KM Compatible between all members of the Storwize family Works with native and externally virtualized storage Fibre Channel and/or IP Replication Network Protocol Flexibility Mirroring types can be used in any combination Environment Flexibility Storage Network Virtual LUNs Node Node Virtual LUNs Node Node Storage Network Virtual LUNs Node Managed Disks Node Virtual LUNs Node Node Virtual LUNs Node Node Virtual LUNs Node Node Virtual LUNs Node Node Virtual LUNs Node Node Managed Disks

24 Spectrum Virtualize REMOTE COPY 24

25 Spectrum Virtualize Metro Mirror Function Metropolitan distance synchronous remote mirroring function Up to 300km between sites for business continuity As with any synchronous remote replication, performance requirements may limit usable distance Host I/O completed only when data stored at both locations Unless write rates are low, not a good candidate for IP based replication Important to keep subsystem performance similar Designed to maintain fully synchronized copies at both sites Once initial copy has completed During resync data is not consistent; use a point-in-time copy All mirroring types/transports delivered as single feature Offers great implementation flexibility Operates between members of the Storwize Family at each site Local and remote volumes may be on any Spectrum Virtualize supported disk systems In order to replication between SVC and Storwize, the receiving Storwize must be in the replication layer 25

26 Spectrum Virtualize Global Mirror Function Long distance asynchronous remote mirroring function Up to km distance (250ms) between sites for business continuity Does not wait for secondary I/O before completing host I/O Helps reduce performance impact to applications Designed to maintain consistent secondary copy at all times Once initial copy has completed Copies are not consistent during a re-sync Use a FlashCopy for a good point-in-time Built on Metro Mirror code base All remote copy functions delivered as single feature Offers great implementation flexibility Operates between Storwize members at each site Local and remote volumes may be on any Storwize Family supported disk systems To replicate between SVC and Storwize, the Storwize system must be in the replication layer 26

27 Low Bandwidth Remote Mirroring GMCV Potential Problem: Spectrum Virtualize Global Mirror delivers very high data currency at remote site but requires network bandwidth to support this currency No mechanism to control RPO Global Mirror with Change Volumes (GMCV) option makes regular copies of data at primary location and transmits them to secondary location RPO and network bandwidth consumption can be regulated User can select frequency of these copies, which determines network bandwidth required and data currency Provides much more flexibility to meet specific application currency and network cost requirements GMCV is limited to 256 relationships per system With RPQ and 7.x code this can be lifted to 2048 relationships per system 27

28 Global Mirror with Change Volumes (GMCV) Initially copy all data from primary volume to secondary volume at point-in-time of when GM relationship started When copy completes a point-in-time copy of the secondary is taken to ensure data consistency Cycle is started again Either immediately or when cycle timer trips Maximum Round Trip Latency FC with DH8, V9K, V7KG2, V5KG2-250 ms. All other equipment 80 ms IP - 1Gb 80ms, 10Gb 10ms Bandwidth sized based on RPO desired (10 minutes to hours) No FlashCopy license is required 28

29 GMCV Cycling Period The user configures the cycling period This is the minimum time between cycle starts Default is 300 seconds (5 minutes) If the link is slow or I/O throughput is high, it may take longer than this for the new data to be transferred to the secondary If so then the next copy will start as soon as the last one has finished RPO is up to 2X the cycling period Actual RPO is determined by the Freeze Time listed on the relationship 29

30 GMCV Cycling Period Options Options for -cyclingmode are none multi Specifies the behavior of Global Mirror for this relationship Specifying the default none gives identical behavior to regular Global Mirror even though you may have change volumes attached to the relationship Specifying multi uses the cycling protocol To start a relationship with cycling mode set to multi there must be change volumes defined for the relationship Admin can create change volumes or let system create them via GUI when configuring GM with Change Volumes option Note that the cycling mode can only be changed when the relationship is stopped and in a stopped state 30

31 Remote Copy Partnership Topologies Up to 4 Spectrum Virtualize family members in direct or indirect partnerships Topologies are related to remote copy partnerships This differs from remote copy relationships (a volume and its mirrored partner) Only 1 IP partnership per Storwize family member is supported The rest must be made up of Fibre Channel relationships. Topology Types Quadrature or Triangle Line Star 31

32 Remote Copy Partnership Topologies Quadrature (or triangle) Partnership Topology Since any system can only have 1 IP partnership only 2 of these partnerships could be IP The rest have to be Fibre Channel Both Storwize systems in this example need to be in the replication layer in this instance If all were Storwize systems, they could be in the default storage layer Copyright IBM Corporation IBM32 Corporation

33 Remote Copy Partnership Topologies Line Partnership Topology Since any system can only have 1 IP partnership only 2 of these partnerships could be IP The rest have to be Fibre Channel All Storwize systems in this example need to be in the replication layer in this instance If all were Storwize systems, they could be in the default storage layer Copyright IBM Corporation IBM33 Corporation

34 Remote Copy Partnership Topologies Star Partnership Topology Since any system can only have 1 IP partnership only 2 of these partnerships could be IP The rest have to be Fibre Channel All Storwize systems in this example need to be in the replication layer in this instance If all were Storwize systems, they could be in the default storage layer Copyright IBM Corporation IBM34 Corporation

35 Fibre Channel Connectivity Options Unix Wintel Max 250 ms roundtrip latency (80ms pre 7.4 and older hardware) SVC/Storwize IP WAN DS5K DS8K Others SVC/Storwize Native Fibre Channel - Including FCoE Dark Fiber FCIP Either Brocade or Cisco DWDM/CWDM DS3/4/5K Others XIV Note: Storwize members have no knowledge of WAN 35

36 Native IP Remote Copy Partnership Enables use of 1Gbit or 10Gbit Ethernet connections using TCPIP without the need for Fibre Channel or FCIP routers for replication Supports all remote copy modes MM and GM GM with Change Volumes preferred mode Covered by normal remote copy license It is not a new replication offering, but rather a new transport versus using a Fibre Channel network WAN Optimizers are NOT supported with this technology Configuration: Automatic path/session configuration via discovery of remote cluster using IP management addresses of both systems in the IP remote copy partnership Configure IP partnership using either 1Gbit or 10Gbit Ethernet ports but not both at same time Currently cannot switch speeds anywhere in end-to-end link other then WAN E.g. cannot connect 10Gbit port to 1Gbit switch and then connect 1Gbit out to WAN E.g All 10Gbit->WAN<-10Gbit or 1Gbit->WAN<-1Gbit CHAP-based authentication supported Includes Bridgeworks SANSlide IP network optimization technology 36

37 Bridgework s SANSlide Optimization (Figure from REDP5023) With TCP/IP, information transfer slows the further you go. This is because of the latency caused by waiting for acknowledgment of each set of packets sent since the next packet set cannot be delivered until the previous one has been acknowledged by the remote system. 37

38 Bridgework s SANSlide Optimization 2 Latency in standard IP network (1Gb link speed): depends on the routing infrastructure depends on distance Latency = Round Trip Time (RTT) for a single packet set 38 (Figure from REDP5023)

39 Bridgework s SANSlide Optimization 3 (Figure from REDP5023) Enhanced parallelism by using multiple virtual connections (VC) that share the same IP links and addresses: When waiting for one VC s ACK, it sends more packets across other VCs If packets are lost from any VC, data will be retransmitted Artificial Intelligence engine adjusts number of VCs, receive window size, and packet size as appropriate to maintain optimum utilization/performance of the links 39

40 Bridgework s SANSlide Optimization 4 Latency in IP network (1Gb link speed): depends on the routing infrastructure depends on distance Latency = Round Trip Time (RTT) for a single packet set 40 (Figure from REDP5023)

41 Remote Copy IP Partnership Requirements If you use IP replication, you must meet the following requirements: TCP ports 3260 and 3265 are used by systems for IP partnership communications These ports must be open on any firewalls Port 3260 is the port used by the systems to initially discover each other Port 3265 is for the actual IP Replication sessions used to transmit data The maximum supported round-trip latency between sites is 80 ms for a 1 Gbps link The maximum supported round-trip latency between sites is 10ms for a 10 Gbps link The recommended method of replication is Global Mirror with Change Volumes The inter-cluster heartbeat traffic consumes 1 Mbps per link The minimum bandwidth requirement for an inter-cluster link is 10 Mbps Maximum throughput is currently restricted based on use of 1gigE or 10gigE ports One 1gigE port could do up to 111 MB/s unidirectional / 190 MB/s bidirectional Two 1gigE ports could do 220 MB/s unidirectional / 325 MB/s bidirectional One 10gigE ports could do up to 200 MB/s unidirectional Two 10gigE ports could do up to 400 MB/s unidirectional 41

42 Native IP Remote Copy Configuring Ethernet Ports Copyright IBM Corporation IBM42 Corporation

43 Native IP Remote Copy Configuring Partnership Link bandwidth setting is configured using Mbits not MBytes and you set this to what the communication link actually can sustain or what is actually allocated for Spectrum Virtualize replication Background copy rate setting is a percentage of the link bandwidth and determines the bandwidth available for initial sync and resyncs or for GMCV Copyright IBM Corporation IBM43 Corporation

44 Native IP Remote Copy Partnership Properties Copyright IBM Corporation IBM44 Corporation

45 Native IP Replication One Link One I/O Group Either Ethernet port 1 or 2 on a node/node canister can be configured for IP replication but not both The remote copy port group on each system includes two IP address, one from each node, and when initially configured the system will establish which one from each site is used This pairing can not be chosen or changed by an administrator In this example H1 and M2 have established a session If H1 node fails then the session between H1 and M2 fails and the system automatically establishes another session between H2 and either M1 or M2 The IP address used for replication on H1 does not failover to H2 Copyright IBM Corporation IBM45 Corporation

46 Native IP Replication One Link Two I/O Groups The remote copy port group on each system includes four IP address, one from each node, and when initially configured the system will establish which one from each site is used If H1 node fails then the session between H1 and M2 fails and the system automatically establishes another session between H2, H3 or H4 and M1, M2, M3 or M4 Copyright IBM Corporation IBM46 Corporation

47 Native IP Replication One Link More Than Two I/O Groups Copyright IBM Corporation IBM47 Corporation

48 Native IP Replication Dual links with all ports active and no standby ports Remote copy port group setup for dual redundant links: Port group 1: two IP addresses, each on a different node in different system (green) Port group 2: two IP addresses, each on a different node in different system (orange) Maximum of 2 simultaneous IP replication sessions at any time Only one active per node, so in this configuration if node A1 would fail then Link 1 would no longer be available and only link 2 would be active Result is loss of half of the bandwidth available for replication If bandwidth sizing dictates dual links required and if one link fails resulting in MM or regular GM not able to operate properly, recommendation is to have 2 I/O Groups at each site to allow the failed link to restart on other nodes available in systems Copyright IBM Corporation IBM48 Corporation

49 Native IP Replication Dual Links - One I/O Group, One Port per Node H1 Primary Volume H2 1 2 Remote Copy Port Group Remote Copy Port Group 2 M1 Secondar y Volume M2 Again maximum of 2 simultaneous IP replication sessions at any time Only one active per node so in this example if node M2 would fail then Link 1 would no longer be available and only link 2 would be active Result is loss of half of the bandwidth available for replication Copyright IBM Corporation IBM49 Corporation

50 Native IP Replication Dual Links - Two I/O Groups, One Port per Node Each remote copy port group on each system includes two IP address, one from I/O Group in this example, and when initially configured the system will establish the pairings used If H1 node fails then the session between H1 and M2 fails and the system automatically establishes another session between H3 and either M2 or M4 since they are all in the same remote copy port group with H1 Copyright IBM Corporation IBM50 Corporation

51 Native IP Replication Miscellaneous Information Cannot today non-disruptively switch from a FC partnership to an IP partnership Must delete relationships and delete partnership and start over basically If currently replicating over FC and can take outage on hosts replicating volumes, we have a method using sync flag on mkrcrelationship SANSlide What it does and doesn t do: It optimizes link utilization, but not latency It uses standard TCP/IP no UDP It doesn t do any encryption, compression or de-duplication Only two systems can be in one IP Replication partnership Can do A->B partnership Cannot do A->B->C->D or A/B/C->D A system can be in an IP partnership with one other system and in FC partnership with others Still a maximum of 4 systems in partnership with each other FC replication and IP replication can be used with any combination of Spectrum Virtualize systems all running V7.2 Copyright IBM Corporation IBM51 Corporation

52 Replication Implementation Considerations Port Usage Although not required, it is highly recommended to dedicate ports to the remote copy process Dedicate SAN ports for FC mirroring Dedicate IP ports for IP replication ISCSI and IP replication not recommended on the same ports Port dedication can help with the following problems: SAN buffer to buffer credit issue for FC replication If ports are shared, this can cause production traffic slowdown 1920 errors due to insufficient WAN bandwidth Spectrum Virtualize Node to Node communication issues On Spectrum Virtualize and V9000 with code 7.x, this will require that you use port masking to take node to node traffic off the dedicated ports The local_fc_port_mask parameter can be set on a cluster to restrict the node to node traffic to specific ports Other members of the Storwize family (V3700,V5000,V7000) do not need this set, because node to node traffic prefers the back-plane to the FC SAN. Copyright IBM Corporation IBM52 Corporation

53 Best Practice Port Designations 53

54 Replication Implementation Considerations What type of partnership do I use? Use an IP partnership when: FCIP equipment is not an option Only when IP network is stable Does not work well on IP networks that are not healthy When bandwidth requirements do not exceed MB/s Only use 10Gb when latency does not exceed 10 ms Use a Fibre Channel partnership when You have reached the maximum (1) IP partnerships per system and need another High bandwidth is a requirement (greater than MB/s) You have a method of taking FC over the WAN (FCIP routers) Copyright IBM Corporation IBM54 Corporation

55 Replication Implementation Considerations What kind of remote copy relationship do I use? Use Metro Mirror when: You absolutely have to have a base RPO of 0 Base RPO - Meaning rolling logs forward does not count Only when you have the bandwidth to sustain it Production traffic will be slowed if you don't Use Global Mirror when: You don't need base RPO of 0, but want a small base RPO (2 min or less) You don't quite have the bandwidth for Metro Mirror You don't want your production applications to be affected by somewhat slower replication Use of 7.2 code can improve situations when bandwidth is not an issue Use Global Mirror with Change Volumes when: You can tolerate a base RPO of 10 minutes or greater Your production applications are being affected by slow downs from Global Mirror Your WAN link is not sufficient to keep up with the peaks from Global Mirror Generally when you are using IP based replication Copyright IBM Corporation IBM55 Corporation

56 Mirroring Adjustable Parameters Link Bandwidth Amount of bandwidth in Mb/s for Spectrum Virtualize to use on the WAN link Background Copy Rate % of link bandwidth that can be used for initial sync and re-sync traffic GMCV is always initial synch/resync traffic Relationship Bandwidth Limit CLI only lssystem, chsystem The speed in MB/s that any ONE relationship is allowed to go GM Link Tolerance CLI only lssystem, chsystem Specified the length of time in seconds for which an inadequate intersystem link is tolerated for a GM operation. Default = 300 seconds GM Max Host Delay CLI only lssystem, chsystem Specifies the maximum time delay in ms, at which the GM link tolerance timer starts counting down. Default = 5 ms 56

57 Mirroring Adjustable Parameters GM Inter-cluster Delay Simulator CLI only lssystem, chsystem Can simulate a delay between Spectrum Virtualize systems in ms GM Intra-cluster Delay Simulator CLI only lssystem, chsystem Can simulate a delay between I/O groups for Hyperswap distance simulation in ms Max Replication Delay CLI only lssystem, chsystem Maximum amount of time in seconds that I/O can be delayed without producing a 1920 error 57

58 Multi-Cloud Storage Gateway Vision Future Capability Hybrid Use Cases: The Future of Storage Economics Spectrum Virtualize Compressed Backup Private Cloud Encrypted Replicated DR Rackspace Tiering Amazon S3 Archive Spectrum Scale Data sharing Integrity Validated Microsoft Azure Multi-cloud Storage Gateway Note: The Multi-cloud Storage Gateway is a planned future enhancement Vision is to support other platforms like Spectrum Accelerate and DS

59 59

60 Legal Notices Copyright 2015 by International Business Machines Corporation. All rights reserved. No part of this document may be reproduced or transmitted in any form without written permission from IBM Corporation. Product data has been reviewed for accuracy as of the date of initial publication. Product data is subject to change without notice. This document could include technical inaccuracies or typographical errors. IBM may make improvements and/or changes in the product(s) and/or program(s) described herein at any time without notice. Any statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. References in this document to IBM products, programs, or services does not imply that IBM intends to make such products, programs or services available in all countries in which IBM operates or does business. Any reference to an IBM Program Product in this document is not intended to state or imply that only that program product may be used. Any functionally equivalent program, that does not infringe IBM's intellectually property rights, may be used instead. THE INFORMATION PROVIDED IN THIS DOCUMENT IS DISTRIBUTED "AS IS" WITHOUT ANY WARRANTY, EITHER OR IMPLIED. IBM LY DISCLAIMS ANY WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE OR NONINFRINGEMENT. IBM shall have no responsibility to update this information. IBM products are warranted, if at all, according to the terms and conditions of the agreements (e.g., IBM Customer Agreement, Statement of Limited Warranty, International Program License Agreement, etc.) under which they are provided. Information concerning non-ibm products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products in connection with this publication and cannot confirm the accuracy of performance, compatibility or any other claims related to non-ibm products. IBM makes no representations or warranties, ed or implied, regarding non-ibm products and services. The provision of the information contained herein is not intended to, and does not, grant any right or license under any IBM patents or copyrights. Inquiries regarding patent or copyright licenses should be made, in writing, to: IBM Director of Licensing IBM Corporation North Castle Drive Armonk, NY U.S.A. 60

61 Information and Trademarks IBM, the IBM logo, ibm.com, IBM System Storage, IBM Spectrum Storage, IBM Spectrum Control, IBM Spectrum Protect, IBM Spectrum Archive, IBM Spectrum Virtualize, IBM Spectrum Scale, IBM Spectrum Accelerate, Softlayer, and XIV are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at The following are trademarks or registered trademarks of other companies. Adobe, the Adobe logo, PostScript, and the PostScript logo are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States, and/or other countries. IT Infrastructure Library is a Registered Trade Mark of AXELOS Limited. Linear Tape-Open, LTO, the LTO Logo, Ultrium, and the Ultrium logo are trademarks of HP, IBM Corp. and Quantum in the U.S. and other countries. Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. Java and all Java-based trademarks and logos are trademarks or registered trademarks of Oracle and/or its affiliates. Cell Broadband Engine is a trademark of Sony Computer Entertainment, Inc. in the United States, other countries, or both and is used under license therefrom. ITIL is a Registered Trade Mark of AXELOS Limited. UNIX is a registered trademark of The Open Group in the United States and other countries. * All other products may be trademarks or registered trademarks of their respective companies. Notes: Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput that any user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the performance ratios stated here. All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions. This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information may be subject to change without notice. Consult your local IBM business contact for information on the product or services available in your area. All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. Information about non-ibm products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot confirm the performance, compatibility, or any other claims related to non-ibm products. Questions on the capabilities of non-ibm products should be addressed to the suppliers of those products. Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography. This presentation and the claims outlined in it were reviewed for compliance with US law. Adaptations of these claims for use in other geographies must be reviewed by the local country counsel for compliance with local laws. 61

62 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 nonibm 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 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 generally-available 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. 62

63 Trademarks The following are trademarks of the International Business Machines Corporation in the United States, other countries, or both. Not all common law marks used by IBM are listed on this page. Failure of a mark to appear does not mean that IBM does not use the mark nor does it mean that the product is not actively marketed or is not significant within its relevant market. Those trademarks followed by are registered trademarks of IBM in the United States; all others are trademarks or common law marks of IBM in the United States. For a complete list of IBM Trademarks, see *, AS/400, e business(logo), DBE, ESCO, eserver, FICON, IBM, IBM (logo), iseries, MVS, OS/390, pseries, RS/6000, S/30, VM/ESA, VSE/ESA, WebSphere, xseries, z/os, zseries, z/vm, System i, System i5, System p, System p5, System x, System z, System z9, BladeCenter The following are trademarks or registered trademarks of other companies. Adobe, the Adobe logo, PostScript, and the PostScript logo are either registered trademarks or trademarks of Adobe Systems Incorporated in the United States, and/or other countries. Cell Broadband Engine is a trademark of Sony Computer Entertainment, Inc. in the United States, other countries, or both and is used under license therefrom. Java and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. Intel, Intel logo, Intel Inside, Intel Inside logo, Intel Centrino, Intel Centrino logo, Celeron, Intel Xeon, Intel SpeedStep, Itanium, and Pentium are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. UNIX is a registered trademark of The Open Group in the United States and other countries. Linux is a registered trademark of Linus Torvalds in the United States, other countries, or both. ITIL is a registered trademark, and a registered community trademark of the Office of Government Commerce, and is registered in the U.S. Patent and Trademark Office. IT Infrastructure Library is a registered trademark of the Central Computer and Telecommunications Agency, which is now part of the Office of Government Commerce. * All other products may be trademarks or registered trademarks of their respective companies. Notes: Performance is in Internal Throughput Rate (ITR) ratio based on measurements and projections using standard IBM benchmarks in a controlled environment. The actual throughput that any user will experience will vary depending upon considerations such as the amount of multiprogramming in the user's job stream, the I/O configuration, the storage configuration, and the workload processed. Therefore, no assurance can be given that an individual user will achieve throughput improvements equivalent to the performance ratios stated here. IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply. All customer examples cited or described in this presentation are presented as illustrations of the manner in which some customers have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics will vary depending on individual customer configurations and conditions. This publication was produced in the United States. IBM may not offer the products, services or features discussed in this document in other countries, and the information may be subject to change without notice. Consult your local IBM business contact for information on the product or services available in your area. All statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. Information about non-ibm products is obtained from the manufacturers of those products or their published announcements. IBM has not tested those products and cannot confirm the performance, compatibility, or any other claims related to non-ibm products. Questions on the capabilities of non-ibm products should be addressed to the suppliers of those products. Prices subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography. 63