IBM Spectrum Protect (TSM) 7.1.3 Technical Overview Tricia Jiang September, 2015.
-
Upload
beverly-ball -
Category
Documents
-
view
370 -
download
16
Transcript of IBM Spectrum Protect (TSM) 7.1.3 Technical Overview Tricia Jiang September, 2015.
© 2015 IBM Corporation
Performance
The performance data contained herein was obtained in a controlled, isolated environment. Actual results that may be obtained in other operating environments may vary significantly. While IBM has reviewed each item for accuracy in a specific situation, there is no guarantee that the same or similar results will be obtained elsewhere.
Customer experiences
Customer experiences described herein are based upon information and opinions provided by the customer. The same results may not be obtained by every user.
IBM’s future plans and Statements of Direction
Any statements regarding IBM's future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
Unannounced products
This document is intended to provide you with information concerning the properties and operational characteristics of products that have not yet been announced by IBM. Although this information has been reviewed for accuracy as of the date of its initial publication, the properties and operational characteristics of the unannounced products discussed in this document are subject to change without notice, and represent statements of IBM's future current intent. This information could include technical inaccuracies or typographical errors. IBM may make improvements and/or changes in the product(s) and/or programs(s) at any time without notice.
Third-party products
Information in this document concerning non-IBM products was obtained from the suppliers of those products, from their published announcements or other generally available sources. Except as otherwise noted, IBM has not tested those products and cannot confirm the accuracy of performance, compatibility, or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the supplier of those products.
Trademarks
IBM, the IBM logo, ibm.com, Tivoli, FlashCopy, Real-time Compression, System Storage, XIV, ProtecTIER, IBM Active Cloud Engine, Storwize, FlashSystem, POWER, and DB2 are trademarks of International Business Machines Corp., registered in many jurisdictions worldwide. Java and all Java-based trademarks and logos are trademarks or registered trademarks of Oracle and/or its affiliates. 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.Other company, product and service names might be trademarks of IBM or other companies. A current list of IBM trademarks is available on the Web at "Copyright and trademark information" at www.ibm.com/legal/copytrade.shtml.
All information contained in this document is subject to change without notice.
The client examples described are presented as illustrations of how those clients have used IBM products and the results they may have achieved. Actual environmental costs and performance characteristics may vary by client.
Information concerning non-IBM products was obtained from a supplier of these products, published announcement material, or other publicly available sources and does not constitute an endorsement of such products by IBM. Sources for non-IBM list prices are taken from publicly available information, including vendor announcements and vendor worldwide webpages. Questions on the capability of non-IBM products should be addressed to the supplier of those products.
All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only.
The actual performance 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 or performance improvements equivalent to the ratios stated here.
Starting price may not include a hard drive, operating system or other features. Prices are subject to change without notice. Contact your IBM representative or Business Partner for the most current pricing in your geography.
The products described in this document are NOT intended for use in implantation, life support, space, nuclear, or military applications where malfunction may result in injury or death to persons. The information contained in this document does not affect or change IBM product specifications or warranties. Nothing in this document shall operate as an express or implied license or indemnity under the intellectual property rights of IBM or third parties.
THE INFORMATION CONTAINED IN THIS DOCUMENT IS PROVIDED ON AN “AS IS” BASIS WITHOUT WARRANTY OR CONDITION OF ANY KIND, WHETHER EXPRESS OR IMPLIED. In no event will IBM be liable for damages arising directly or indirectly from any use of the information contained in this document.
References in this document to IBM products or services do not imply that IBM intends to make them available in every country.
© IBM Corporation 2014. All rights reserved2
Trademarks
© 2015 IBM Corporation3
Spectrum Protect (TSM) 7.1.3 Announce 9/1/2015 GA 9/11/2015
TSM for VE - DP for VMware Self service File Restore
TSM for ERP - DP for SAP HANA Port to Power Linux - SLES 11 (Big Endian)
FCM Unix/LinuxILMT v4 compliance
Pre-installation checker enhancements
Oracle RAC on ASM cloning
Scripted capacity reporting for TSM SUR and SUR FE
FCM WindowsExchange 2013 IMR support for Recoverable Items folder (Dumpster)
Scripted capacity reporting for TSM SUR and SUR FE
FCM VMwareScripted capacity reporting for TSM SUR and SUR FE
Dynamic allocation of SVC target volumes for FCM for VMware only
SERVERNextGen Dedupe
Native Object Storage for SoftLayer/Swift
Encryption at rest for Object Storage
Embedded component currency - DB2, Gskit, Install Manager, JazzSM, ITM
ILMT v4 Compliance
OPERATIONS CENTERNextGen Dedupe wizard for creating storage pools
Object Store Integration
Replication visibility and management
Decommission node
Policy management enhancements
Front end Capacity Reporting
New Theme
CLIENT ILMT v4 compliance (Oracle, SAP for DB2, SAP for Oracle, SAP HANA)
B/A client support of GPFS 4.1.1
TSM Client Installation improvements - Eliminate separate language packs
FB4WCurrency, minor enhancements, security & APAR fixes
Covered in a separate presentation
© 2015 IBM Corporation
Server - NextGen Dedupe and Container Storagepools
Container storagepools– Storage is handled in an automatic fashion such that no direct management of the storage is required
• Philosophy - write once and don’t fuss with it• NO reclamation, migration, copies, backups, shredding, LAN-free…..• NO device classes or volumes like legacy random or sequential access storagepools
– Dynamic creation & deletion
Next generation deduplication (NextGen dedupe) uses container storagepools– Allows for the deduplication of all data from both the client and server
• With the exception of files that have been encrypted on the client side– Deduplication occurs in-line
• As opposed to legacy TSM server dedupe being done in a separate process following ingest
Goals of container and NextGen dedupe are to provide– Easier management– Faster performance– More scalability
4
© 2015 IBM Corporation
Server – NextGen Dedupe -vs- Legacy Dedupe
5
Server – NextGen Dedupe -vs- Legacy Dedupe
5
Server side NextGen dedupe in container stgpool
Server side legacy dedupe in file dev stgpool
Client side dedupe with container stgpool
Client side dedupe with legacy file dev stgpool
When is data deduped Server side Inline Out-of-Band: Dedupe processing & reclamation required after backups finish
Before client sends data to server
Before client sends data to server
Data transferred from client to server Fully hydrated data Fully hydrated data Dehydrated data Dehydrated data
Where is data stored Container Stgpool File Dev Class Stgpool Container Stgpool File Dev Class Stgpool
Storage pool encryption supported Yes in cloud containers No Yes in cloud containers No
Client side encryption supported Yes (not deduped) Yes (not deduped) No (SSL works with dedupe) No (SSL works with dedupe)
“protect stgpool”(Replicate data only)
Yes in directory containers No Yes in directory containers No
“repair stgpool” Yes in directory containers No Yes in directory containers No
“replicate node” (Replicate data & meta data)
Yes Yes Yes Yes
“replicate node recoverdamaged=yes” No Yes No Yes
Stgpool reclamation required No Yes No Yes
TSM version supported 7.1.3+ 6.1+ 7.1.3+ 6.2+
Special license needed No No No No
Amount of data processed per day 80 TB per day 20 TB per day 100 TB per day 30 TB per day
Data managed on a single TSM server 1 to 4 PB 100 to 400 TB 1 to 4 PB 100 to 400 TB
Max physical capacity on a TSM server 1 PB 100TB 1 PB 100TB
© 2015 IBM Corporation
Server - Container Storagepools
Container Locations chosen by TSM admin during setup– Directory Container
• File based local storage using file system directories– Cloud Container (On-Premises or Off-Premises)
• Object based IBM® SoftLayer, V1Swift or OpenStack Swift cloud computing system• Only difference in on-premises or off-premises, is who is managing the cloud and what type of cloud
Container sub-types created dynamically by TSM – Dedup container– Non-dedupe container
• Data that does not de-dupe like encrypted data• Data that shouldn’t be de-duped like client meta-data
How containers are utilized by TSM– Container sizes currently based on file system size– As extents are deleted, space is freed up in containers for new incoming client data– TSM re-uses existing containers if there is room in the container and the data object being written fits– Containers deleted when
• Stgpooldir containing the container is updated or destroyed• No more references to the data because of expiration or deletion of filespace
– TSM server might run defragmentation on the container• MOVE CONTAINER is used to manually defragment a directory container
Operations container stgpools do not support– Storage pool backup, migration, aggregation, reclamation, simultaneous write, active data pools, NDMP dedupe
streams, data shredding, storage pool CRCs, DRM functions, identify duplicates, LAN Free6
© 2015 IBM Corporation
Server – Container Storagepools – Input/Output Refinements (IOR)
A combination of technologies to improve IO throughput, allowing customers to achieve greater quantities of data protection– Asynchronous IO Writes; Direct IO; In Line Deduplication– Wisdom to know when to apply the appropriate solution
No customer configuration capabilities, it dynamically adjusts to load demands– Data flows from a client session request to storage – Synchronous or asynchronous IO path
Deduplication– Prior to writing data, the inline deduplication engine identifies if the chunk needs to be written– If the chunk is already on disk the chunks reference count is merely incremented
When the unit of work is small– The chunk will be written using synchronous IO– The overhead of asynchronous IO would result is slower performance
When the unit of work is large – Chunk write requests will be broken up into work requests – Chunk dispatched into a producer/consumer based asynchronous IO handler– Additional test in asynchronous IO for in-flight chunks prior to final write request
Write to Filesystem– Once it is determined where a chunk is to be written, the data is flushed to the storage system using buffered IO or direct IO– The database will be updated upon successful completion of the chunk write– Finally, success or failure (and request for retry) will be communicated back to the session
7
© 2015 IBM Corporation
Server – NextGen Dedupe and Container Storagepools - Extents
Extents What extents are:
– Pieces of data stored in a storage pool– Can be composed of metadata or file data– Can be deduplicate-able or not– Can vary in size – Sliding scale window optimized during ingest
What extents are not:– Whole objects unto themselves
• Objects require a list of extents for reconstruction
What extents do:– The means for deduplicating and storing data– Allows for easy matching of data – already stored? Discard it!
What is a non-deduplicated extent?– Data TSM can’t deduplicate on the server is a non-deduplicated extent
Enable Server-side Inline data de-duplication Client side:
– DEDUPLICATION No– Optional: RESOURCEUTILIZATION >1
Server side:– DEFINE STGPOOL <pool name> STGTYPE=CONTAINER– UPDATE COPYGROUP <domain_name> <policy_set_name> <class_name> DESTINATION=<container stgpool>
8
© 2015 IBM Corporation
Server and OC – Container Storagepools
Wizard in OC can help define container storage pools– In OC: Storage / Storage Pools / +Storage Pool
Directory-container pool data– Need fully-qualified path name
• For example, c:\temp\dir1\ for Microsoft Windows or /tmp/dir1/ for UNIX• Directory must exist with read/write privileges
– Distributing data across the available directories can improve I/O performance• Only if the storage pool directories map to different physical disks
– Restriction:• Do not use a directory on same physical fs as the server database and logs• Storage pool directories can grow to use more space on the file system
– Directory-containers must have overflow pools• Overflow pool can be a cloud pool or a primary pool on disk or tape• Can’t be another directory based stgpool
Cloud-Containers – Requires connection information that the server uses to access the cloud when sending and retrieving storage pool data– Currently supports SoftLayer and OpenStack Swift, V1Swift clouds– Currently supports platforms AIX, xLinux, Solaris, Windows– By default, the storage pool has read/write access
• Can change this setting with UPDATE STGPOOL – By default, encrypt, using server encryption, storage pool data in off-premises cloud-container pools
• Optionally encrypt data in on-premises cloud-container pools9
© 2015 IBM Corporation
Server - Native Object Storage for SoftLayer/Swift
Wikipedia's definition of Object Storage: – “Object Storage (also known as object-based storage) is a storage architecture that manages data as objects, as
opposed to other storage architectures like file systems which manage data as a file hierarchy and block storage which manages data as blocks within sectors and tracks.”
Cloud containers are mostly based on the directory containers – The same deduplication and chunking of client side objects applies– TSM writes each chunk to a container in the cloud instead of a container in a directory
TSM creates containers in the cloud automatically– The customer does not need to create the containers – To use the cloud TSM just needs the parameters specified in the DEFINE STGPOOL command
• URL, user name, and password for the cloud
Writing to a cloud involves I/O over a network as well as many other external factors that influence performance– A smaller network pipe or a OpenStack Swift environment that is underpowered, creates very slow results in TSM – Object Storage may not be right for all use cases
10
© 2015 IBM Corporation
Server and OC – Container Storagepools – Protect Stgpool
Protect STGPool Schedule Replicates backup and archive data to the target server without the associated metadata
– Data can be transferred more quickly than with client replication alone– Improves replication performance– Enables the repair of damaged files on the source server
For directory-container pools, when replication is enabled
Requires replication schedule to replicate metadata– Until node replication runs and the metadata is synchronized, data that was transferred by storage pool protection
cannot be restored from the target server
Should run often enough to keep pace with incoming data– Should complete before node replication starts
“protect stgpool” command• forcereconcile - Reconciles the chunks between the two servers
• Anything not protected will be protected• Anything that is on the target, but not on the source will be deleted on the target
• purgedata=yes - removes all replicated chunks on target• purgedata=no - default, runs normal protect stgpool• purgedata=deleted –
• Does not protect any new chunks• Only removes chunks from target that are no longer on source
• maxsessions=2-100 - number of data sessions that can send data to a target server• preview=yes/no – preview the data• wait=yes/no - wait for the server to process this command in the foreground
11
© 2015 IBM Corporation
Server – Container Storagepool - Repair Storagepool
• Repairs deduplicated extents in a directory-container storage pool• Replaces data that has been damaged on the source server with valid data obtained from the target
• Requirements:• Protect stgpool must be setup and have been run to the target• Target server must be available • Target storage pool must not be damaged • Network must be available
• REPAIR STG <Stgpool>
• Parameters:• MAXSESSions=1-20 – number of data sessions that can send data to a target server• Preview=yes/no – preview the data• Wait=yes/no - wait for the server to process this command in the foreground
12
© 2015 IBM Corporation
Server – Container Storagepools - Defragmentation
Defragmentation – defragmentation is to directory container storagepools, as reclamation is to legacy storagepools
Quarantining Directory Containers– When containers are full, they are not eligible for writing and are quarantined– Allows for performance increases by avoiding choosing containers that aren't optimal
Defrag analysis – Occurs every hour on quarantined containers (sooner if deletions are occurring)– Assigns a FRAGLEVEL to the quarantined containers
• FRAGLEVEL is a percentage of how many additional average sized extents could be stored after container defragmentation is performed• If originally 2 additional extents could be stored, but after defrag 5 could be stored, the FRAGLEVEL is 60
– After analysis, container stays in quarantine until re-analyzed• If FRAGLEVEL <= 5
Defragmenting Directory Containers– After container quarantined and analyzed, candidate for defrag
• If FRAGLEVEL >= 80 – Not an in-place defragmentation– Requires a new container to be created and all extents copied to the new container
Releasing containers from quarantine– After container analyzed, candidate for release from quarantine
• If FRAGLEVEL < 80 (too low for defrag) and FRAGLEVEL > 5 (too high for quarantine) • If 10% of the container is free, regardless of the FRAGLEVEL
Query Fragmentation - find information on the status of fragmented containers Related commands
– Query Fragmentation– Move container
13
© 2015 IBM Corporation
Server – Container Storagepools
define stgpooldirectory - Creates a directory for container storage pools Create the directory on the file system first
define stgpool stgtype=container - creates a directory container
define stgpool stgtype=cloud - creates a cloud container
query container – shows container stgpool properties
move container – manually defragment a directory container
audit container – Sees if an audit is needed by checking the active log
query damaged pool_name type=container – Displays information about damaged files
repair stgpool –Repair deduplicated extents in a container storage pool• Must run PROTECT STGPOOL so there is a copy for use in repair
protect stgpool – date in directory-container replicated to target container stgpool
14
© 2015 IBM Corporation
Sever and OC – NextGen Dedupe - Statistics
Storage Pool Level • How efficient deduplication is working on a broad storage pool level • QUERY STGPOOL <pool name> FORMAT=DETAILED (Column: “Deduplicate Data Not Stored”)• SELECT STGPOOL_NAME,SPACE_SAVED_MB FROM STGPOOLS• Operations Center %saving
Node Level • How efficient deduplication is working on data provided by a single node• GENERATE DEDUPSTATS • QUERY DEDUPSTATS• DELETE DEDUPSTATS
• Records are kept indefinitely• Each record is one row in a new table• Can use Delete to remove one of the records
Session Level• How efficient deduplication worked on an individual session• SELECT * FROM SUMMARY
15
© 2015 IBM Corporation
Server – Encryption at Rest for Object Storage
Server encrypts data before it writes it to the cloud container storage pool Storage pool encryption was designed to require minimal user interaction – "Set it and forget it"
Storage Pool Encryption currently limited to cloud containers– Administrators want additional protection for data stored outside of their control– Encryption is enabled by default for OFF-PREMISES cloud pools, is optional for ON-PREMISES cloud pools
Encryption is enabled at a storage pool level– From cmdline, specify ENCRYPT=YES|NO on DEFINE and UPDATE STGPOOL commands
• Yes: client data is encrypted by the server • No: client data is not encrypted by the server
– Once encryption is enabled, all new data stored into the pool will be encrypted, existing data is not affected
Node Replication Considerations– Long term goal is to make node replication aware of encryption, but not currently….
• Encrypted chunks are decrypted when read by replication• Chunks are (re)encrypted on target as necessary, based on target storage pool settings• Recommend using SSL between source and target servers
16
Server Encryption Client Encryption
Server controls encryption at storage pool level Client controls encryption at the object level using include.encryption and encryptiontype options
Server manages all of its encryption keys Client optionally stores keys in tsm.pwd file
Server uses AES encryption with minimum key length of 256-bits Clients uses AES128, AES256 or DES56 bit
Server signs encrypted data to detect tampering or corruption Client does not
Data encrypted by the server can be deduplicated Data encrypted by the client cannot be deduplicted
© 2015 IBM Corporation
Server – Encryption at Rest for Object Storage
Encryption keys are managed at dedup chunk level– A unique, random encryption key is created for each chunk and is used to encrypt the data
Master encryption key is used to encrypt each chunk key– Master encryption key is automatically generated following upgrade to 7.1.3 or for new 7.1.3 instance– Stored outside of database in server password file dsmserv.pwd in server instance directory
• Encrypted (obfuscated) using key derived from server password (SET SERVERPASSWORD)
Server password must be set in order to create the master encryption key• Permissions grant access only to user who runs the server (typically the instance user)• On Windows, members of Administrators group also have access to allow server to run as a service
Protecting encryption keys– When encrypting data, it is critically important to protect the master encryption key
• Lose the master encryption key, and you lose access to all of your encrypted data– Database backup has been enhanced to optionally back up the master encryption key
• DB BACKUP already regularly used to protect the server instance• Optional because a password must be specified to protect the master encryption key on media• Even though parameters are optional, stgpool encryption cannot be configured unless DB Backup is first
configured to automatically back up the master encryption key• SET DBRECOVERY ... PROTECTKEYS=YES PASSWORD=pwdtext
– In the event of a disaster, master encryption key can be recovered using RESTORE DB
Server side encryption adds some additional time to the overall backup times– On the order of 1-2% of the total time required to do backups– Time spent writing to the cloud is a couple of orders of magnitude larger than equivalent disk I/Os
17
© 2015 IBM Corporation
Operations Center – Encryption at Rest for Object Storage
Part of Operations Center’s Define stgpool wizard– In OC: Storage / Storage Pool / +Storage Pool
ENCRypt Specifies whether the server encrypts client data before it writes it to the storage pool– Yes: client data is encrypted by the server – No: client data is not encrypted by the server– Default depends on the physical location of the cloud, which is specified by the CLOUDLOCATION parameter
• If the cloud is off premise, the OC wizard selects encryption by default• If the cloud is on premise, the server does not encrypt data by default
Data can be encrypted at rest (at the Protect server) before it goes to the container • SET DBRECOVERY must be set to use cloud encryption
• In OC: Servers / Details / Properties • Keys are managed by the Protect server
• Dbbackup protects encryption key• In OC: Servers / Back Up
• Master key protects cloud encryption keys• Needed to restore encrypted pool!
18
© 2015 IBM Corporation
Server – Container Storagepools
19
Directory Container
On-Premises Cloud Container
Off-Premises Cloud Container
Storage pool encryption Not Available Optional Default Option
Over-flow pool Legacy Stgpool or cloud container
Not Available Not Available
Replicate node Optional Optional Optional
Protect stgpool Optional Not available Not Available
Repair stgpool Yes, if protect stgpool run
No No
Storage mechanism File Based Object BasedSwift, V1Swift, Softlayer
Object BasedSwift, V1Swift, Softlayer
Move container Yes No No
Off-line reorgs for Inventory tables Yes Yes Yes
Defrag container Yes No No
Audit container Yes Yes Yes
© 2015 IBM Corporation
Server – Online Index Reorganization
DB2 on-line index reorganization utilizing “cleanup only” index reorganization• Without noticeable performance degradation• Without excessive active log usage• The same space reduction as off line reorganization• Index space use, over time, is more stable and able to reuse existing space, versus continually growing
Can remove from TSM server options file, as these options are now ignored– DISABLEREOGINDEX– DISABLEREORGCLEANUPINDEX
Do not include in the TSM server options file– ALLOWREORGTABLE– ALLOWREORGINDEX
Include in the TSM server options file (default no tables)– DISABLEREORGTable BF_AGGREGATED_BITFILES,BF_BITFILE_EXTENTS,BACKUP_OBJECTS,ARCHIVE_OBJECTS
Tech Note: http://www-01.ibm.com/support/docview.wss?uid=swg21683633
20
© 2015 IBM Corporation
Operations Center – New Theme Updated Overview and most other panels in the OC Reflects the "flatter" style that is the current direction of GUIs Panels are much more "responsive“
– They don't fall apart when you resize the browser window – Allows you to shrink browser
21
resize
© 2015 IBM Corporation
Operations Center – Object Store Integration
Status of Container storage pools listed
Bytes saved and Deduplication Ratio listed
New Table Columns: Container Type, %Savings, Deduplication Enabled (hidden by default)
22
© 2015 IBM Corporation
Operations Center – Replication ‘+Server Pair’ Wizard
23
Operations Center / Storage / Replication / + Server Pair• Both TSM Servers must be at 7.1.3• Both the source and target must be part of the monitored server list • Walks through setting up replication
• Will be prompted to setup or enter server password
• Sets replication for all nodes on source• Can exclude client or set of clients from being replicated
• In OC: Clients / More / Set Replication• Disabled nodes will be skipped• Nodes with name conflicts will be overwritten and sync'd (with warning displayed first)
• Does not sync policies• Much smoother going if the target server has the same policies as the source server
• Schedules replication node• Schedules protect stgpool if directory pool present
© 2015 IBM Corporation
Operations Center – Replication Visibility and Management
24
In Operations Center: Storage / Replication1. Workload: • The amount of data that must be transferred to synchronize the source and target server• Stripped portion is = amount of data that has already been processed during the currently running schedule2. Triangle: • The amount of data that is projected to be replicated during this schedule3. Yellow bracket: • Many to one replication relationship• Primary servers Electric, Helix, Ion and Nucleus are replicating to the target server Fusion4. Gray Bracket: • Unavailable server• Server Molecule is unavailable currently5. Red and black circle: • Server is unavailable for replication processing6. Question Mark in a diamond: • Unknown server, the OC is still obtaining server info7. Down arrow in a circle: • The server is running an earlier SW version8. Workload direction: • Replication completion trend• Comparison of yesterday’s workload with previous day• Synchronized, Catching up, steady, falling behind
© 2015 IBM Corporation
Operations Center – Replication Visibility and Management
25
In Operations Center: Storage / Replication / Details• Workload
• Expands to show details for 5 tasks• Link to additional tasks
• 2 Week history• Workload in dark blue• Backed up in light blue• Replicated in purple
© 2015 IBM Corporation
Operations Center – Decommission NodeUtilize if a client node is no longer needed in the production environment
“decommission node” initiates a gradual, controlled decommission – In OC: Clients / More / Decommission
Removes a client node from the production environment – Node’s backup data expires according to policy settings
• Unless you explicitly delete the data– All backup file versions are made inactive copies– After all data retention periods expire, deletes the node’s file spaces
• Equivalent to the DELETE FILESPACE command
This action cannot be reversed and causes deletion of data – Cannot recommission the client node – Client node cannot access the server – Client node’s data is not backed up– Client node is locked and cannot be unlocked– Node’s file spaces are eventually removed– Deletes all schedule associations for the node
• Schedules are no longer run on the client node – Prevents the client from accessing the server
• Equivalent to LOCK NODE command
Must have system, unrestricted policy, or restricted policy privilege for the policy domain to which the client node belong “Q Node f=d & Q Filespace f=d” command shows if decommissioned “decommission vm” – removes an individual virtual machine within a data center node “deactivate data” – only applies to Oracle db client nodes
26
© 2015 IBM Corporation
Operations Center – Policy Management Enhancements In OC: Services / Details / Policy Sets Edit policy sets
View and edit policysets Changes highlighted in yellow, click “-” to delete
Add new management classes Activate management class (changes in yellow, deletions in red)
27
© 2015 IBM Corporation
Operations Center – Front End Capacity Reporting
Capacity usage info is collected automatically every 7 days– Collection interval cannot be modified– BA Client, DP/Domino, DP/Exchange, DP for VMware, DP
for Hyper-V, DP/Oracle, DP/SQL, DP/SAP for DB2, DP/SAP for Oracle, DP/SAP HANA, TSM for Space Management for UNIX and Linux
Capacity estimates are for planning purposes only
Front-end– Estimates the capacity of IBM TSM SUR - Front End – Based on size of primary data that is protected for clients– Licensing depends on amount of data that must be recovered to
restore the protected files to the latest recovery point– Must be TSM 7.1.3 or later
Back-end– Estimates the capacity usage of TSM for SUR – Based on amount of managed data in primary storage pools– Copy & active-data pool data are copies of data in primary
storage pools and are not counted in the measurement– Deduplication and compression can reduce costs– Must be TSM 7.1.1 or later
28
© 2015 IBM Corporation
Operations Center – Client Diagnosis using Client Management Services
In Operations Center: Clients / Details / Diagnosis
Enables TSM admins to view & understand client problems without having to log into the client machine
Client Management Service (CMS) software runs on the client machine itself, responding to requests from the TSM OC
Obtain installation package from product DVD or IBM download site– 7.1.3.000-TIV-TSMCMS-Linuxx64.bin– 7.1.3.000-TIV-TSMCMS-Windows32.exe– 7.1.3.000-TIV-TSMCMS-Windows64.exe
Changes Since 7.1.1– Shifted from OSLC to REST with JSON payload– Allows 3rd Party REST API (future), and more consumable by integrators– Some minor changes in CmsConfig commands/syntax to simplify configuration– 7.1.3 OC center won’t work with 7.1.1 CMS code
29
© 2015 IBM Corporation
API – Operations Center Rest API
REST API has access to most of the same information as Operations Center – Utilizes either JSON or XML format– Enables customers to display the data as they want to see it
Admin REST API lives on the hub server
Enable in the Operations Center using the Settings dialog, by default off
•Get functions
•Put functions
•Post functions
30
© 2015 IBM Corporation
Server & Client – ILMT v4 Compliance
IBM License Metric Tool (ILMT) enablement is done to comply with an internal process– Intended to be "transparent" to the end TSM / IBM Spectrum Protect customer
7.1.3 introduces a new ILMT Enablement Guide (covers 6.4. – 7.1.3)– For users who must identify, collect, maintain & report information about the inventory of IBM software assets by using
ILMT– Identifies ILMT enablement position of each product for TSM 6.4.0 - 7.1.2 & IBM Spectrum Protect 7.1.3 family of products– Available on 7.1.3 GA, in a tech note: http://www.ibm.com/support/docview.wss?uid=swg21961752
Questions regarding how to use ILMT software should be redirected to ILMT Support– General information about the IBM Passport Advantage licensing options is available at:
http://www.ibm.com/software/passportadvantage/about_software_licensing.html
– The License Agreement (LA) and License Information (LI) documents that are provided with IBM Spectrum Protect and TSM define the terms and conditions that are associated with usage of that software
– The LI document defines the licensing option (also known as the charge metric) that is associated with the licensed offering
31
© 2015 IBM Corporation
Client – Spectrum Scale (GPFS) 4.1.1 support
Currency– IBM Spectrum Scale (former GPFS) 4.1.1 support for backup and Space Management
• Linux x86_64 requirements page has exact Spectrum Scale version supported with 7.1.3• http://www-01.ibm.com/support/docview.wss?uid=swg21248771
White paper : Support for the Spectrum Scale function "AFM (Active File Management)“– Publishing after 7.1.3 is announced– Covers the major user scenarios for AFM (disaster recovery, branch office, system migration)
• Describes the setup of TSM in such environments• Lists the challenges and recommendations for such a setup
– 7.1.3 books were updated and will link the white paper– White paper will be linked to this page:
https://www.ibm.com/developerworks/community/wikis/home?lang=en#!/wiki/Tivoli%20Storage%20Manager/page/Integration%20Scenarios%20Tivoli%20Storage%20Manager%20Server%20and%20Tivoli%20Storage%20Manager%20Server%20Extended%20Edition
New Function on Spectrum Scale side:– Support for the Spectrum Scale function "snapshot based backup of filesets“ added– TSM backup archive option "SNAPSHOTROOT" improved to support this function – Now possible to backup a single Spectrum Scale fileset stand alone with mmbackup
32
© 2015 IBM Corporation
FB4W – Currency, minor enhancements, security & APAR fixes
Spectrum Protect for Workstations 7.1.3 (FastBack for Workstations / CDP) Upgrades:
– Java 8 (Java 8.0.1/Java 8 SR1)– Liberty Server 8.5.5.5– Upgrade to InstallShield 2014 SP1 – Upgrade GSKit to v8.0.50.44 – Upgrade TSM API to v7.1.3
New Browser Support Client– Firefox/Mozilla V29 - 36– Google Chrome v34 - 41
New Browser Support Central Administration Console– Firefox/Mozilla V31 ERS, V34 - V36
Client– Executable & msi files are signed indicating IBM as the publisher– Fix for security vulnerability for Logjam and Bar Mitzvah attacks
Central Administration Console– Fix for install of CAC service, previously might get stuck at 'starting' and not get into 'running' state– Executable & msi files are signed indicating IBM as the publisher
33
© 2015 IBM Corporation
Hands-On Labs
Enablement Virtual Delivery Platform (EVDP) has 4 Hands-On Labs available for education and demo of Spectrum Protect 7.1.3 as of Announce:
Business Partners and IBM Internal technical staff can request access by sending an e-mail to Michele Payne at [email protected]
vApp Name Description
Lab Protect VE 713 Self Service File This short hands-on lab covers configuration of Spectrum Protect for Virtual Environments and use of the Self-Service File Restore on Windows.
Lab Protect 713 Dedup Container Storage This series of exercises covers configuration and usage of Spectrum Protect container storage pools with inline deduplication.
Lab Protect 713 OC, Node Rep, FB4W These 2 vms are used for the Spectrum Protect 7.1.3 Node Replication, Operation Center and Fast Back for Workstation labs. The OC and Node Rep both have two sections: one covering what is new in 7.1.3, and one covering the functions basics. The student can choose to do one or both of the sections for either lab.
© 2015 IBM Corporation
References7.1.3 webcasts:– IBM Spectrum Protect Sales Update
• http://www-304.ibm.com/services/weblectures/dlv/partnerworld/ltu50665– Technical Update - What's New in Spectrum Protect 7.1.3 - Server, BA Client and OC
• IBM: http://lt.be.ibm.com/stg/ltu50747• BP: http://www.ibm.com/services/weblectures/dlv/partnerworld/ltu50747
– Technical Update - What's New in Spectrum Protect 7.1.3 - Virtual Environments and Snapshot• IBM: http://lt.be.ibm.com/stg/ltu50749• BP: http://www.ibm.com/services/weblectures/dlv/partnerworld/ltu50749
– Skills Development Transfer: TTEC GO – deep dive training• http://learning.atlanta.ibm.com/hr/global/events/ems.nsf/pages/event-sessions?open&id=ECB34EB871EE7A5085257EAB008379F4
Hands on Labs:– Enablement Virtual Delivery Platform will continue to add hands-on labs
• Spectrum Protect Snapshot 4.1.3 with Exchange, SQL Server, and File Systems• Spectrum Protect Snapshot 4.1.3 with DB2• Spectrum Protect Rest API• Spectrum Protect Operation Center• Spectrum Protect Node Replication
Youtube videos:– Subscribe to Enablement’s YouTube Channels
• Spectrum Storage Software Training
https://www.youtube.com/channel/UCGkjRNkO0AQNyQbWhS1tTzw• Tricia Jiang
https://www.youtube.com/user/TSMevangelist• Sean Sperry
https://www.youtube.com/user/seansperry