Computer Drive User Manual
Table Of Contents
- Front cover
- Contents
- Notices
- Preface
- Summary of changes
- Part 1 Overview
- Chapter 1. Introduction
- Chapter 2. Copy Services architecture
- Part 2 Interfaces
- Chapter 3. DS Storage Manager
- Chapter 4. DS Command-Line Interface
- Chapter 5. System z interfaces
- Part 3 FlashCopy
- Chapter 6. FlashCopy overview
- Chapter 7. FlashCopy options
- 7.1 Multiple relationship FlashCopy
- 7.2 Consistency Group FlashCopy
- 7.3 FlashCopy target as a Metro Mirror or Global Copy primary
- 7.4 Incremental FlashCopy - refresh target volume
- 7.5 Remote FlashCopy
- 7.6 Persistent FlashCopy
- 7.7 Data set FlashCopy
- 7.8 Reverse restore
- 7.9 Fast reverse restore
- 7.10 Options and interfaces
- Chapter 8. FlashCopy ordering and activation
- Chapter 9. FlashCopy interfaces
- Chapter 10. FlashCopy performance
- Chapter 11. FlashCopy examples
- Part 4 Metro Mirror
- Chapter 12. Metro Mirror overview
- Chapter 13. Metro Mirror options and configuration
- Chapter 14. Metro Mirror interfaces
- 14.1 Metro Mirror interfaces - overview
- 14.2 TSO commands for Metro Mirror management
- 14.3 ICKDSF
- 14.3.1 Metro Mirror management with ICKDSF
- 14.3.2 Display the Fibre Channel Connection Information Table
- 14.3.3 PPRCOPY DELPAIR
- 14.3.4 PPRCOPY DELPATH
- 14.3.5 PPRCOPY ESTPATH
- 14.3.6 PPRCOPY ESTPAIR
- 14.3.7 PPRCOPY FREEZE
- 14.3.8 PPRCOPY QUERY
- 14.3.9 PPRCOPY RECOVER
- 14.3.10 PPRCOPY SUSPEND
- 14.3.11 PPRCOPY RUN
- 14.3.12 Refreshing the VTOC
- 14.4 DS Command-Line Interface
- 14.5 DS CLI command- examples
- 14.6 DS Storage Manager GUI
- 14.7 ANTRQST API
- Chapter 15. Metro Mirror performance and scalability
- Chapter 16. Metro Mirror examples
- Part 5 Global Copy
- Chapter 17. Global Copy overview
- Chapter 18. Global Copy options and configuration
- Chapter 19. Global Copy performance and scalability
- Chapter 20. Global Copy interfaces
- Chapter 21. Global Copy examples
- Chapter 22. Global Mirror overview
- Part 6 Global Mirror
- Chapter 23. Global Mirror options and configuration
- 23.1 Terminology used in Global Mirror environments
- 23.2 Create a Global Mirror environment
- 23.3 Modify a Global Mirror session
- 23.4 Remove a Global Mirror environment
- 23.5 Global Mirror with multiple storage disk subsystems
- 23.6 Connectivity between local and remote site
- 23.7 Recovery scenario after primary site failure
- 23.7.1 Normal Global Mirror operation
- 23.7.2 Primary site failure
- 23.7.3 Failover B volumes
- 23.7.4 Check for valid Consistency Group state
- 23.7.5 Set consistent data on B volumes
- 23.7.6 Reestablish the FlashCopy relationship between B and C volumes
- 23.7.7 Restart the application at the remote site
- 23.7.8 Prepare to switch back to the local site
- 23.7.9 Return to local site
- 23.7.10 Conclusions
- Chapter 24. Global Mirror interfaces
- 24.1 Global Mirror interfaces - overview
- 24.2 Different interfaces for the same function
- 24.3 Global Mirror management using TSO commands
- 24.3.1 Establish a Global Mirror environment
- 24.3.2 Define paths
- 24.3.3 Establish Global Copy volume pairs
- 24.3.4 Establish FlashCopy relationships for Global Mirror
- 24.3.5 Define a Global Mirror session
- 24.3.6 Populate a Global Mirror session with volumes
- 24.3.7 Start a Global Mirror session
- 24.3.8 Query a Global Mirror session
- 24.4 DS CLI to manage Global Mirror volumes in z/OS
- 24.5 Global Mirror management using ICKDSF
- 24.5.1 Establish a Global Mirror environment
- 24.5.2 Define paths
- 24.5.3 Establish Global Copy pairs
- 24.5.4 Establish FlashCopy relationships
- 24.5.5 Define a Global Mirror session
- 24.5.6 Add volumes to a session
- 24.5.7 Start Global Mirror
- 24.5.8 Query an active Global Mirror session
- 24.5.9 Remove a Global Mirror environment
- 24.5.10 Stop the Global Mirror session
- 24.5.11 Remove volumes from Global Mirror
- 24.5.12 Un-define the Global Mirror session
- 24.5.13 Withdraw FlashCopy relationships
- 24.5.14 Delete Global Copy pairs
- 24.5.15 Remove all paths
- 24.6 ANTRQST macro
- 24.7 DS Storage Manager GUI
- Chapter 25. Global Mirror performance and scalability
- Chapter 26. Global Mirror examples
- 26.1 Global Mirror examples - configuration
- 26.2 Global Mirror query examples with TSO
- 26.3 Set up the Global Mirror environment using TSO
- 26.4 Primary site failure and recovery management with TSO
- 26.4.1 Primary site failure
- 26.4.2 Stop a Global Mirror session
- 26.4.3 Failover from B to A volumes
- 26.4.4 Check Global Mirror FlashCopy status between B and C volumes
- 26.4.5 Create a data consistent set of B volumes
- 26.4.6 Optionally create a data consistent set of D volumes
- 26.4.7 Create a data consistent set of C volumes
- 26.4.8 Prepare to return to the local site
- 26.4.9 Replicate the changes from B to A
- 26.4.10 Return to the local site and resume Global Mirror
- 26.5 Remove Global Mirror environment using TSO
- 26.6 Planned outage management using ICKDSF
- 26.7 Remove a Global Mirror environment using ICKDSF
- 26.8 Query device information with ICKDSF
- 26.9 Set up a Global Mirror environment using DS SM
- 26.10 Set up a Global Mirror environment using the DS CLI
- 26.11 Control and Query Global Mirror with the DS CLI
- 26.12 Site switch basic operations using the DS CLI
- 26.13 Remove the Global Mirror environment with the DS CLI
- Part 7 Interoperability
- Chapter 27. Combining Copy Service functions
- Chapter 28. Interoperability between DS6000 and DS8000
- 28.1 DS6000 and DS8000 Copy Services interoperability
- 28.2 Preparing the environment
- 28.2.1 Minimum microcode levels
- 28.2.2 Hardware and licensing requirements
- 28.2.3 Network connectivity
- 28.2.4 Creating matching user IDs and passwords
- 28.2.5 Updating the DS CLI profile
- 28.2.6 Adding the Storage Complex
- 28.2.7 Volume size considerations for Remote Mirror Copy
- 28.2.8 Determining DS6000 and DS8000 CKD volume size
- 28.3 RMC: Establishing paths between DS6000 and DS8000
- 28.4 Managing Metro Mirror or Global Copy pairs
- 28.5 Managing DS6000 to DS8000 Global Mirror
- 28.6 Managing DS6000 and DS8000 FlashCopy
- 28.7 z/OS Global Mirror
- Part 8 Solutions
- Chapter 29. Interoperability between DS6000 and ESS 800
- 29.1 DS6000 and ESS 800 Copy Services interoperability
- 29.2 Preparing the environment
- 29.2.1 Minimum microcode levels
- 29.2.2 Hardware and licensing requirements
- 29.2.3 Network connectivity
- 29.2.4 Creating matching user IDs and passwords
- 29.2.5 Updating the DS CLI profile
- 29.2.6 Adding the Copy Services domain
- 29.2.7 Volume size considerations for RMC (PPRC)
- 29.2.8 Volume address considerations on the ESS 800
- 29.3 RMC: Establishing paths between DS6000 and ESS 800
- 29.4 Managing Metro Mirror or Global Copy pairs
- 29.5 Managing ESS 800 Global Mirror
- 29.6 Managing ESS 800 FlashCopy
- Chapter 30. IIBM TotalStorage Rapid Data Recovery
- Chapter 31. IBM TotalStorage Productivity Center for Replication
- 31.1 IBM TotalStorage Productivity Center
- 31.2 Where we are coming from
- 31.3 What TPC for Replication provides
- 31.4 Copy Services terminology
- 31.5 TPC for Replication terminology
- 31.6 TPC for Replication session types
- 31.7 TPC for Replication session states
- 31.8 Volumes in a copy set
- 31.9 TPC for Replication and scalability
- 31.10 TPC for Replication system and connectivity overview
- 31.11 TPC for Replication monitoring and freeze capability
- 31.12 TPC for Replication heartbeat
- 31.13 Supported platforms
- 31.14 Hardware requirements for TPC for Replication servers
- 31.15 TPC for Replication GUI
- 31.16 Command Line Interface to TPC for Replication
- Chapter 32. GDPS overview
- Appendix A. Concurrent Copy
- Appendix B. SNMP notifications
- Appendix C. Licensing
- Appendix D. CLI migration
- Related publications
- Index
- Back cover

336 IBM System Storage DS6000 Series: Copy Services with IBM System z
25.1 Performance aspects for Global Mirror
Global Mirror is basically comprised of Global Copy and FlashCopy, and it combines both
functions to create a solution that provides consistent data at a distant site. Global Copy has,
at most, only a minimal impact on the response time of an application write I/O to the Global
Copy primary volumes.
In the Global Mirror environment, FlashCopy is used with the
nocopy attribute. This implies
that, for any write I/Os to the FlashCopy source volume, there will be additional internally
triggered I/Os in the remote storage disk subsystem. These I/Os preserve the FlashCopy
source volume tracks by making a copy to the FlashCopy target volume
before the source
tracks are updated. This happens every time within the interval in between two FlashCopy
establish operations.
Note that the nocopy attribute does not start a background copy operation from source to
target. It only maintains a set of FlashCopy bitmaps for the source and target volumes. These
bitmaps are established the first time that the FlashCopy relationship is created with the
nocopy attribute. Before a source track is modified between the creation of two Consistency
Groups, the track is copied to the target volume to preserve the previous point-in-time copy.
This includes updates to the corresponding bitmaps to reflect the new location of the track,
which belongs to the point-in-time copy. Be aware that each Global Copy write to a secondary
volume in the window of two adjacent Consistency Groups causes such a FlashCopy I/O
operation.
For an application I/O within this context, there is also a Global Copy I/O when Global Copy
replicates a track from the primary volume to the secondary volume. This implies an
additional read I/O in the primary storage disk subsystem and a corresponding write I/O to the
secondary storage disk subsystem. In a Global Mirror environment, the Global Copy
secondary volume is also the FlashCopy source volume. This might have some effect in a
very busy disk subsystem.
Figure 25-1 approximates what happens in between two Consistency Group creation events.
The application write I/O completes immediately at the local site (1). Global Copy reads the
modified track at the local site (2). Before the track gets updated on the B volume, the
FlashCopy source track is preserved on the C volume because of the nocopy attribute. This is
a read (3) and a write (4) which are done to preserve the source track writing it to the C
volume. Finally, the track is updated on the B volume by another write (5). This is not
necessarily the exact sequence of internal I/O events, but rather an approximation. There are
optimization and consolidation effects that make the entire process more efficient.
Figure 25-1 Application write I/O within two Consistency Group formation events
Primary
Primary
A
Primary
PENDING
Write
Host
FICON
A1
Read
FCP links
Primary
Primary
A
Secondary
PENDING
B1
Primary
Primary
Tertiary
C1
Write
Read
Write
1
2
3
4
5
Remote site
Local site