IBM SONAS Enterprise backup and remote replication solution in a private cloud

IBM SONAS Enterprise backup and
remote replication solution in a private
cloud

Gaurav Chhaunker

IBM Systems and Technology Group ISV Enablement
January 2012

© Copyright IBM Corporation, 2012

Table of contents
Abstract........................................................................................................................................1
Executive summary ....................................................................................................................1
Private cloud and Active Cloud Engine ....................................................................................1
Prerequisites ...............................................................................................................................2
Introduction to IBM SONAS .......................................................................................................2
IBM SONAS architecture ............................................................................................................3
Introduction to Tivoli Storage Manager ....................................................................................5
Overview of Active Cloud Engine ..............................................................................................6
SONAS – Active Cloud Engine integration ...............................................................................7
Active Cloud Engine: Caching modes in SONAS ....................................................................8
Overview of the solution and usecases ....................................................................................8
Setting up the application and manual changes.................................................................................... 10
Data back up at home site using Tivoli Storage Manager ..................................................................... 19
Initial data migration to cache site using Active Cloud Engine pre-population and policy engine......... 20
Synchronization of data at home and cache site using Snapshots ....................................................... 23
Restoration of data at the home site...................................................................................................... 25
Two-way replication (home to remote and back ).................................................................................. 27
From home site to cache site .......................................................................................... 27
From Cache site to home site ......................................................................................... 27
Summary....................................................................................................................................30
Acknowledgement ....................................................................................................................31
Appendix A: Resources............................................................................................................32
Appendix B: Glossary...............................................................................................................33
About the authors .....................................................................................................................35
Trademarks and special notices..............................................................................................36

IBM SONAS Enterprise backup and remote replication solution in a private cloud

Abstract
In a globally dispersed enterprise with private cloud environment, where unstructured data is
exponentially growing, there is a need to provide 24x7 accesses to business-critical data and be
able to restore in case of loss of data. IBM Scale Out Network Attached Storage (IBM SONAS)
with its integrated IBM Tivoli Storage Manager client enables enterprises to back up and restore
data seamlessly and the IBM Active Cloud Engine offers the capability to replicate data to
remote sites.
This paper provides an end-to-end solution for users using an application, such as Apple Final
Cut Pro (a video editing s/w) and one of the currently supported protocols from SONAS, for
example Network File System (NFS). This user data is then replicated to other remote (cache)
sites using Active Cloud Engine. The data is also backed-up using Tivoli Storage Manager at the
home (primary) site. The paper also explains the various steps in setting up and configuring
Active Cloud Engine.

Executive summary
Small and large enterprises continue to demand storage solutions that can store massive amounts of file-
based data in data centers across geographies with ease of management and can scale on demand.
Often enterprises with fast-growing file systems face limitations of scalability and performance with
traditional network-attached storage (NAS) filers because of the requirement to work on millions and
billions of active files in parallel. IBM® SONAS is a multipetabyte scale-out NAS storage offering for
information storage. It is designed to scale out to store substantial amount of active files with superior
performance and ease of management.
IBM Tivoli® Storage Manager Client is fully integrated in IBM SONAS. The integrated solution enables
enterprises to back up and restore data in the least possible time. IBM SONAS with Tivoli Storage
Manager provides a backup and restore solution to the organizations thus increasing efficiency,
performance, and reliability. While taking a full back up every time ensures reliability of backups, it also
supports incremental backups.
IBM SONAS built-in IBM Active Cloud Engine™ feature helps to selectively distribute files to local or
remote sites globally for sharing, data protection, content distribution, and so on. Active Cloud Engine
enables consolidation by centralizing all files under a single file system view, enables multiple SONAS
systems across the globe to be connected and presented to the user as a single namespace. It enables
local access to files from across the globe quickly and efficiently. Active Cloud Engine also helps to create
powerful and scalable storage environment and improve data protection by identifying candidates for
backup or disaster recovery.

Private cloud and Active Cloud Engine
Applications built on cloud architectures are such that the underlying computing infrastructure is used only
when it is needed (for example to process a user request), draw the necessary resources on demand
(such as compute servers or storage), perform a specific job, then relinquish the unnecessary resources
and often dispose themselves after the job is done. While in operation, the application scales up or down
elastically based on resource needs.

1

One of the definitions of private clouds is, in situations where activities and functions are provided as a
service over a company's intranet. Private clouds are built by an organization for its own users, and
everything is delivered within the organization's firewall (instead of the Internet). The private cloud owner
does not share resources with any other company, and therefore multi-tenancy is not an issue.
By totally owning a cloud computing environment, an enterprise can provide and govern computing
resources (such as physical servers, application servers, storage space, applications, services, and so on)
in an efficient, compliant, and secure manner. At the same time, by using a private cloud, an enterprise
can also achieve significant cost saving from the infrastructure's consolidation and virtualization.
In a private cloud environment, Active Cloud Engine helps to alleviate some of the challenges of a private
cloud. The policy engine helps to seamlessly and efficiently manage data, which is distributed across the
globe. Active Cloud Engine is tightly integrated with IBM General Parallel File System™ (IBM GPFS). The
different caching modes of Active Cloud Engine enable the organizations to customize their internal
collaboration. Also, Active Cloud Engine ensures that users always access the latest and updated file and
the same is changed at the remote site, and only incremental changes are sent on the wire, thereby
optimally using the network resources and reducing the cost. Active Cloud Engine creates a single
namespace for its users.

Prerequisites
The following are the minimum requirements to be able to set up Active Cloud Engine in a private cloud
environment.
 Minimum two SONAS with same version Release 1.3 or higher
([root@sonas1isv.mgmt001st001 ~]# get_version; Node Version: 1.3.0.0-74f)
 SONAS should be reachable (ping) from each other
 All the nodes and CTDB-status in SONAS clusters should be healthy (command to check,
cli – lsnode)

In addition, the paper assumes familiarity with the following concepts.
 Basic knowledge of IBM SONAS
 Basic knowledge of Tivoli Storage Manager

Introduction to IBM SONAS
The demand to manage and store massive amounts of data continues to challenge the cloud environment.
For that reason, IBM has designed a new IBM SONAS solution to embrace the cloud storage and the
petabyte (PB) age. It can meet today’s storage challenges with quick and cost-effective, IT-enabled
business enhancements that can grow with unprecedented scale. It can also deliver computing services
that make the technology, underlying user devices, almost invisible. It enables applications and services to
be uncoupled from the underlying infrastructure, enabling businesses to adjust quickly to change. As a
result, IBM SONAS can easily integrate with an organization’s strategies to develop a more dynamic
enterprise.

Applications and services such as digital media, healthcare, defense and government, interactive games,
and web content are driving enormous data usage in the industry. This is increasingly generating more
requirements for architectural scalability and I/O response times than ever before. All these industries
deploying network-attached storage (NAS) environment face challenges, such as:

2

 Tremendous growth of unstructured data with same (or less) administrative staff
 Current NAS solutions do not scale both in performance and capacity
 No parallel access to data
 Backup windows are too small for hundreds of terabytes (TB) of data
 No easy way to apply policies across independent filers
 No policies to automatically migrate data
 Difficulty in globally implementing a centrally-managed and centrally-deployed automatic-
tiered storage, for example, information lifecycle management (ILM) that is essential when
TB multiply into PB

IBM SONAS helps to overcome all these limitations with its rich features.

IBM SONAS architecture
IBM SONAS, when attached to a customer’s Internet Protocol (IP) network, serves files using industry-
standard protocols, such as NFS, Common Internet File System (CIFS), File Transfer Protocol (FTP),
Hypertext Transfer Protocol (HTTP), and Hypertext Transfer Protocol Secure (HTTPS). You can use
SONAS for storage needs ranging from 100 TB to 21 PB.

3

Figure 1: SONAS architecture

As shown in Error! Reference source not found., the unique architecture of IBM SONAS offers greater
flexibility when compared to traditional NAS solutions. Depending on the kind of workload involved, you
can choose to go for multiple interface nodes or storage pods.

Element Function

Integrated interface / Management nodes Performs both management and console tasks for
system as well as allows clients to connect to the
system for file based services and asynchronous
replication function

Fast InfiniBand® network Interconnects management, interface, and storage
nodes

Private management network Interconnects all physical components of the system

4

One or more 42U Enterprise Racks Places all the elements of the system

Optional IBM Tivoli Storage Manager Performs hierarchical storage management (HSM)
nodes and backup services for the system

Table 1: SONAS system components

Introduction to Tivoli Storage Manager
IBM Tivoli Storage Manager is a client/server program that provides centralized, automated data
protection, and storage management solutions to customers in a multivendor computer environment. Tivoli
Storage Manager provides a policy-managed backup, archive, and space-management facility for file
servers, workstations, applications, and application servers.
Tivoli Storage Manager includes the following components:

 Server − Server program
− Administrative interface
− Server database and recovery logs
− Server storage

 Client nodes − Backup-archive clients
− Network-attached storage file server
− Application client
− Application program interface (API)

 Tivoli Storage Manager for space management
 Storage agents

Network clients use Tivoli Storage Manager to store data for any of the following purposes:

 Backup and restore
 Archive and retrieve
 Instant archive and rapid recovery
 Migrations and recall

Through Tivoli Storage Manager server, you can manage the devices and media used to store client data.
The server integrates the management of storage with the policies you define for managing client data.

5

Overview of Active Cloud Engine
Active Cloud Engine is a scalable, high performance remote file data caching solution integrated with
GPFS and the single global namespace features of SONAS. In Active Cloud Engine terminology, the
nodes where file system operations originate leading to Active Cloud Engine operations are termed as
application or compute nodes. Active Cloud Engine remote operations are ran on one or more GPFS
nodes called gateway nodes, which have connectivity to remote nodes in home cluster. All the nodes
become compute nodes by default, and no special designation required to be assigned. Figure 2 shows a
high level view of the Active Cloud Engine concepts.

Figure 2: Overview of Active Cloud Engine

The caching functionality is enabled by defining an Active Cloud Engine file set at the caching site with a
caching relationship to the file set at the home site exported over NFS.
All the gateway nodes do the NFS mount of the exported path from the home site. A request on the
application node is satisfied from cache if the data is already cached. If the data is not in cache, it sends
the request to the gateway node. The gateway node then brings data over the WAN and stores it in the
local GPFS file system from where application node accesses it directly.
Active Cloud Engine also offers disconnected mode operations, where if gateway nodes are
disconnected from home cluster then data already cached continues to be served by the application
nodes. Cached data can be configured to expire after being in the disconnected state for a certain amount
of time. This prevents access to stale data, where staleness is defined as the amount of time, cache is out
of synchronization with the data at home or as the amount of time, cache is in disconnected state from
home.
Active Cloud Engine performs whole file caching, where data is brought into the cache even when the file
is read partially. However, read requests are split among various gateway nodes for better performance
and caching of data happens asynchronously.

6

SONAS – Active Cloud Engine integration
The integration work aims to integrate and leverage Active Cloud Engine into SONAS for WAN caching
features by mapping Active Cloud Engine concepts onto the SONAS environment.
Figure 3 gives a high-level view on mapping GPFS Active Cloud Engine onto SONAS. Some of the
SONAS interface nodes acts as Active Cloud Engine gateway nodes. A SONAS configuration CLI is
provided to configure Active Cloud Engine within SONAS by defining gateway node designations. Data
access in SONAS happen through interface nodes, hence they act as application nodes also. It is
possible to configure a dedicated network between the home and cache cluster for WAN data transfer.

Figure 3: SONAS – Active Cloud Engine integration

The various steps followed at the cache cluster for pulling data from home cluster, if data isn’t cached
already are:
1. NAS client issues a read request to SONAS cache cluster through the SONAS interface node.
2. If data is not present in the cache, the request is forwarded to the SONAS wcache node.
3. The wcache node then sends a request to the home cluster through NFS mounts from the home
cluster.
4. The wcache node receives data in response.
5. The wcache node stores the data received from home into the local cache cluster storage
persistently.
6. The SONAS interface node which was waiting for this process till now, reads the data from the
local cache cluster storage.

7

7. The SONAS interface node serves data to the NAS client.

Active Cloud Engine: Caching modes in SONAS

User can create an Active Cloud Engine file set in different modes. The following three modes are
supported in SONAS.

 Single-writer: An Active Cloud Engine file set in this mode has exclusive write permission
to the data which is present at home. No other cache cluster or NAS clients connected to
the home cluster would be able to write to the data at home cluster, that is, only one cache
cluster can be configured as single-writer mode. It avoids conflicts between the data at the
cache cluster and home cluster.
SONAS restricts only one cache site to be configured as single-writer through the CLI
commands: mkwcachesource and mkwcache.

 Local-updates: An Active Cloud Engine file set will be allowed to read data from home
server but any writes done to the cache will not be synchronized back to the home server.
This is similar to a sandbox or scratch pad environment.

 Read-only: An Active Cloud Engine file set will be permitted to read data. Any form of
write operations will fail.

Overview of the solution and usecases
For example, consider a Media and Entertainment enterprise, using Apple Final Cut Pro 10.0.1 for creating
(reading and writing ) and editing media files ( typically *.mov files).
Figure 4 illustrates a typical enterprise environment, which is spread across geographies, where users are
authenticated through Microsoft® Active Directory server and access SONAS as backend for storage and
backup of their day-to-day data.

8

Figure 4: A typical enterprise setup, where users write to the home site

Figure 5: Users writing to cache site after some manual changes in application

9

Setting up the application and manual changes
In order to be able to use the Apple Final Cut Pro and Active Cloud Engine features, the following steps
need to be performed on the Mac system and Final Cut Pro. The user application can interact with SONAS
using any of the SONAS supported protocols, however, the internal communication through Active Cloud
Engine happens over NFS.

1. Click Finder  Go  Connect to Server to mount the SONAS NFS export. Specify the SONAS
home IP address and path.

Figure 6: Connecting to NFS export

Figure 7: SONAS export connection in progress

2. After successfully mounted, the export is displayed in the Finder window.

Figure 8: Mac Finder, showing the mounted export

10

Figure 9: SONAS export on Mac desktop

3. Start the Final Cut Pro application.

Figure 10: Final Cut Pro main screen

4. Click File  Add SAN Location.

11

Figure 11: Adding SAN location in Final Cut Pro

5. In the Select a SAN Location Folder window, the SONAS export is displayed. Double-click the
SONAS export and click Add.

Figure 12: Selecting the SONAS home export

12

Figure 13: Adding the SONAS home export

6. The newly added SAN location is now updated in the main window of Final Cut Pro. Right-click the
SONAS export and click New Event.

Figure 14: The newly added export seen in the main screen

13

Figure 15: The home export displayed in the main window with the menu options

You can now create and edit files as per requirement. The newly created files are also reflected in the
Mac Finder window.

Figure 16: Media files displayed in Mac Finder

In the event of home or primary site outage, after few manual changes in the Final Cut Pro application,
users can now use the cache site with the same files now being available.
Perform the following steps to configure the Final Cut Pro to now mount and use remote (cache) site:

14

Figure 17: Users switching to the cache site after few modifications in application

1. Click Finder  Go  Connect to Server to mount the SONAS NFS export. Specify the SONAS
cache IP address and path.

Figure 18: Mounting the remote site export in Mac

15

Figure 19: Connection to export in progress

After successful mounting, the remote export is displayed on the Mac desktop.

Figure 20: The remote export as seen on Mac desktop

2. Click File  Add SAN Location to add the new SAN location in Final Cut Pro.

16

Figure 21: Adding the remote export in Final Cut Pro SAN location

3. Double-click the cache export.

Figure 22: Selecting the remote export

17

4. As the replication was set (refer to “Initial data migration to cache site using Active Cloud Engine
pre-population and policy engine” section) the file (data) that was created on the home site is now
displayed on the cache site as well. Click Add.

Figure 23: Adding the remote export

The new SAN location is now displayed in the Final Cut Pro main window.

Figure 24: Remote export displayed in the Final Cut Pro main window

The files can now be imported and editing can continue.

18

Figure 25: The files from home site replicated to cache site, now used by the application

Data back up at home site using Tivoli Storage Manager
Tivoli Storage Manager Client software is preinstalled on SONAS on each of the interface nodes. This is
done during the SONAS software installation. Not all the interface nodes need to participate in the backup
and restore activities. All the interface nodes which are configured for backup purposes are called backup
nodes. First backup node which is configured with Tivoli Storage Manager Server is called the primary
backup node. Other backup nodes are called as helper backup nodes.
Perform the following steps to back up data at home site.

1. Verify the Tivoli Storage Manager nodes configuration by using the lstsmnode command.
[root@sonas1isv.mgmt001st001 ~]# lstsmnode
Node name TSM target node name TSM server name TSM server address TSM node name
int004st001 sonas1 isvp13_lpm 9.11.83.12 sonas1int04
EFSSG1000I The command completed successfully.

2. Validate the configured backup file system and backup scheduled task on the SONAS system.
[root@sonas1isv.mgmt001st001 ~]# lsbackupfs
File system TSM server List of nodes
gpfs1 isvp13_lpm int004st001,int005st001,int006st001


3. Invoke the file system backup using the following command. The time taken to back up depends
on the size of the file system.
[root@sonas1isv.mgmt001st001 ~]# startbackup gpfs1
EFSSG0543I A backup of file system gpfs1 started with JobID 0.

4. Verify the back-up log
[root@sonas1isv.mgmt001st001 ~]# showlog 0
Primary node: int004st001
Job ID : 0
PID:1431277
Primary backup node: int004st001
Helper backup nodes: int005st001,int006st001
2011-12-19 02:31:19-07:00 Starting backup.
mmbackup: ignoring unrecognized TSM query output:
Accessing as node: SONAS1
stat: cannot stat `/ibm/gpfs1/mmbackup.audit.gpfs1': No such file or directory

19

--------------------------------------------------------
Backup of /ibm/gpfs1 begins at Mon Dec 19 02:31:23 MST 2011.
--------------------------------------------------------
Mon Dec 19 02:31:30 2011 Could not restore previous shadow file from TSM server isvp13_lpm
Mon Dec 19 02:31:30 2011 Querying files currently backed up in TSM server:isvp13_lpm.
Mon Dec 19 02:31:31 2011 Query of TSM server isvp13_lpm returned 12
Mon Dec 19 02:31:31 2011 No inventory found in TSM server isvp13_lpm
Mon Dec 19 02:31:31 2011 Created empty shadow for first backup to TSM server: isvp13_lpm
Mon Dec 19 02:31:32 2011 Generating policy rules file: /var/mmfs/mmbackup/.mmbackupRules.gpfs1
Mon Dec 19 02:31:33 2011 Scanning file system gpfs1
Mon Dec 19 02:31:33 2011 Determining file system changes for gpfs1 [isvp13_lpm].
Mon Dec 19 02:31:33 2011 Finished calculating lists [19 changed, 0 expired] for server isvp13_lpm.
Mon Dec 19 02:31:33 2011 Sending files to the TSM server [19 changed, 0 expired].
Mon Dec 19 02:44:30 2011 Policy returned 0 Highest TSM error 0
TSM Summary Information:
Total number of objects inspected: 22
Total number of objects backed up: 22
Total number of objects updated: 0
Total number of objects rebound: 0
Total number of objects deleted: 0
Total number of objects expired: 0
Total number of objects failed: 0
Mon Dec 19 02:44:32 2011 Done working with files for TSM Server: isvp13_lpm.
Mon Dec 19 02:44:32 2011 Completed successfully exit 0
----------------------------------------------------------
Backup of /ibm/gpfs1 completed successfully at Mon Dec 19 02:44:33 MST 2011.
----------------------------------------------------------
2011-12-19 02:44:33-07:00 INFO: backup successful (rc=0).
----------------------------------------------------------
End of log - backup completed
----------------------------------------------------------


For more detailed information on how to backup and restore files using Tivoli Storage Manager on
SONAS, refer to the references at the end of paper.

Initial data migration to cache site using Active Cloud Engine pre-population and
policy engine
Usually, the cache (remote) site fetches data from home on a need basis. However, when setting up the
home and remote sites for the first time, typically there is a requirement to move the relatively huge initial
data to cache site. This is achieved using the pre-population feature of Active Cloud Engine. Currently it is
the only way to transfer the initial data.
You need to perform the following steps to setup pre-population for initial data transfer across sites.

At the home site:
1. Create a regular file set. You can also use an existing file set .
[root@sonas1isv.mgmt001st001 ~]# mkfset gpfs1 home_src1
EFSSG0070I File set home_src1 created successfully.

2. Link a regular file set to a junction path.
[root@sonas1isv.mgmt001st001 ~]# linkfset gpfs1 home_src1
EFSSG0015I Refreshing data.
EFSSG0078I File set home_src1 successfully linked.

20


3. Check for the successful creation of a regular file set; you can write / enter data into this file set
using the cp command or use any application for the same
[root@sonas1isv.mgmt001st001 ~]# lsfset gpfs1
ID Name Status Path Is independent Creation time Comment Timestamp
0 root Linked /ibm/gpfs1 yes 12/12/11 4:09 AM root fileset 1/10/12 12:56 AM
1 home_src1 Linked /ibm/gpfs1/home_src1 no 1/10/12 12:56 AM 1/10/12 12:56 AM

4. Create an Active Cloud Engine home export in read-only mode. Here the export is in read-only
mode, and therefore, when users are trying to access the files from cache / remote site, they can
only read and not edit. However, by creating the NFS export in the rw mode will allow to edit the
files.
[root@sonas1isv.mgmt001st001 ~]# mkwcachesource homeshare1 /ibm/gpfs1/home_src1 --client '9.11.82.90(ro)'

5. Check for the successful creation of Active Cloud Engine home export.

[root@sonas1isv.mgmt001st001 ~]# lswcachesource
WCache-Source Name WCache-Source Path ClientClusterId ClientClusterName WCache-Source Access Mode Is
Cached
homeshare1 /ibm/gpfs1/home_src1 12402779243272901319 sonas3isv.storage.tucson.ibm.com ro no

6. Create an NFS export for the application. This is the export which will be used as part of Add SAN
Location in the Apple Final Cut Pro video editing software
[root@sonas1isv.mgmt001st001 ~]# mkexport rwshare1 /ibm/gpfs1/home_src1 --nfs '*(rw,no_root_squash,insecure)'
EFSSG0019I The export rwshare1 has been successfully created.

7. Confirm the successful creation of the export.
[root@sonas1isv.mgmt001st001 ~]# lsexport
Name Path Protocol Active Timestamp
rwshare1 /ibm/gpfs1/home_src1 NFS true 1/10/12 1:00 AM

At the remote site:
1. Configure few interface nodes on cache site as Gateway Nodes.
[root@sonas3isv.mgmt001st001 ~]# mkwcachenode --nodelist int001st001,int002st001

2. Check for the creation of Gateway nodes. The Is Cache column for the selected interface nodes
should be yes
[root@sonas3isv.mgmt001st001 ~]# lsnode -v -r
Hostname IP Description Role Product version Connection status GPFS status CTDB status Username Is
manager Is quorum Daemon ip address Daemon version Is Cache Recovery master Monitoring enabled Ctdb ip address OS name
OS family Serial number Last updated
int001st001 172.31.132.1 interface 1.3.0.0-74f OK active active root yes no
172.31.132.1 1211 yes no yes 172.31.132.1 RHEL 6.1 x86_64 Linux 78998P3 1/10/12 1:04
AM

21

172.31.132.2 1211 yes no yes 172.31.132.2 RHEL 6.1 x86_64 Linux 78998R5 1/10/12 1:04
AM
int003st001 172.31.132.3 interface 1.3.0.0-74f OK active active root yes yes
172.31.132.3 1211 no no yes 172.31.132.3 RHEL 6.1 x86_64 Linux 78998P1 1/10/12 1:04
AM
172.31.132.4 1211 no yes yes 172.31.132.4 RHEL 6.1 x86_64 Linux 78998T0 1/10/12 1:04
AM
172.31.132.5 1211 no no yes 172.31.132.5 RHEL 6.1 x86_64 Linux 78998N4 1/10/12 1:04
AM
172.31.132.6 1211 no no yes 172.31.132.6 RHEL 6.1 x86_64 Linux 787271F 1/10/12 1:04
AM
mgmt001st001 172.31.136.2 active management node management,interface 1.3.0.0-74f OK active active root
yes no 172.31.136.2 1211 no no yes 172.31.136.2 RHEL 6.1 x86_64 Linux KQ977R4
1/10/12 1:04 AM
strg001st001 172.31.134.1 storage 1.3.0.0-74f OK active root no yes
172.31.134.1 1211 no no no 127.0.0.1 RHEL 6.1 x86_64 Linux 78977M5 1/10/12 1:04 AM
strg002st001 172.31.134.2 storage 1.3.0.0-74f OK active root no yes
172.31.134.2 1211 no no no 127.0.0.1 RHEL 6.1 x86_64 Linux 78977M4 1/10/12 1:04 AM

3. Create a read-only Active Cloud Engine file set.
root@sonas3isv.mgmt001st001 ~]# mkwcache gpfs1 remotefset1 /ibm/gpfs1/remotefset1 --cachemode read-only --homeip
9.11.82.16 --remotepath /ibm/gpfs1/home_src1

4. Check for the successful creation of Active Cloud Engine file set.
[root@sonas3isv.mgmt001st001 ~]# lswcache gpfs1
ID Name Status Path CreationTime Comment RemoteFilesetPath CacheState CacheMode
18 remotefset1 Linked /ibm/gpfs1/remotefset1 1/10/12 1:23 AM 9.11.82.28:/ibm/gpfs1/home_src1 enabled read-only

5. The initial data transfer (prepopulation) can be configured based on the policy. You can specify
the file selection through the prefetch policy. Here, all the files from a storage pool are selected.
[root@sonas3isv.mgmt001st001 ~]# mkpolicy prefetchpolicy -R "RULE 'allfiles' LIST 'files' FROM POOL system WHERE NAME LIKE
'%'"

6. Check for the creation of a policy
[root@sonas3isv.mgmt001st001 ~]# lspolicy
Policy Name Declarations (define/RULE)
default default
prefetchpolicy allfiles

7. Start the transfer. This is an asynchronous operation.
[root@sonas3isv.mgmt001st001 ~]# runprepop gpfs1 remotefset1 prefetchpolicy

8. Check the status of prepopulation.
[root@sonas3isv.mgmt001st001 ~]# lsprepop -l 1
Cluster ID filesystem FilesetName Status Last update Timestam Message
12402779243272901319 gpfs1 remotefset1 RUNNING 1/10/12 1:30 AM RUNNING

[root@sonas3isv.mgmt001st001 ~]# lsprepop -l 1

22

Cluster ID filesystem FilesetName Status Last update Timestam Message
12402779243272901319 gpfs1 remotefset1 FINISHED 1/10/12 1:30 AM FINISHED

9. The initial contents from the home site are now seen on the remote site.

[root@sonas3isv.mgmt001st001 ~]# ll /ibm/gpfs1/remotefset1
total 15120
-rw------- 1 root root 15734851 Jan 10 01:11 clip-2011-12-21 14;16;39.mov
drwxrwxrwx 3 root root 8192 Jan 10 01:30 Final Cut Events

Synchronization of data at home and cache site using Snapshots
Active Cloud Engine provides the peer-snapshot (psnap) utility. Snapshot is a point-in-time copy of Active
Cloud Engine file set. Peer snapshots are used for backup and disaster recovery purposes. The psnap
utility takes a snapshot of cache and pushes all the data in cache to the home site so that the data at the
home site is consistent with cache and then takes a snapshot of home. This results in a pair of snapshots
at both sides referring to the same consistent copy of data.
Psnap can be taken for a single-writer mode file set only. It is currently not supported for read-only or
local-update modes. Also, currently the limit of maximum number of snapshots is 100 per file set.
The various steps involved are as follows:

At the home site:
1. Create a file set named cloud_nfs_src2.
root@sonas1isv.mgmt001st001 ~]# mkfset gpfs1 cloud_nfs_src2
EFSSG0070I File set cloud_nfs_src2 created successfully.

2. Link the file set to the /ibm/gpfs1/cloud_nfs_src2 path.
[root@sonas1isv.mgmt001st001 ~]# linkfset gpfs1 cloud_nfs_src2
EFSSG0078I File set cloud_nfs_src2 successfully linked.

3. Create a WAN cache source in the rw mode.
[root@sonas1isv.mgmt001st001 ~]# mkwcachesource homeshare2 /ibm/gpfs1/cloud_nfs_src2 --client '9.11.82.90(rw)'

At the remote site:
1. Create a corresponding WAN cache in the single writer mode.

[root@sonas3isv.mgmt001st001 remotefset]# mkwcache gpfs1 remotefset2 /ibm/gpfs1/remotefset2 --cachemode single-writer --
homeip 9.11.82.16 --remotepath /ibm/gpfs1/cloud_nfs_src2

2. Add data to remotefset2.
[root@sonas3isv.mgmt001st001 remotefset2]# ls -lrt
total 0
-rw------- 1 root root 5242880000 Dec 16 03:13 sample1.mov

23

3. Take a peer snapshot, snap1.
[root@sonas3isv.mgmt001st001 /]# mkpsnap gpfs1 remotefset2 snap1
If the maximum limit of psnaps is reached, this will automatically delete the oldest snapshot.
Do you really want to perform the operation (yes/no - default no):yes
EFSSG0019I The psnap snap1 has been successfully created.

4. List snap1.
[root@sonas3isv.mgmt001st001 /]# lspsnap gpfs1 -r
Filesystem name Fileset name Snapshot ID Status Creation ID Timestamp
gpfs1 remotefset2 snap1 Valid 12/16/11 3:22 AM 1 12/16/11 3:23 AM

5. Verify the contents of snap1.
[root@sonas3isv.mgmt001st001 /]# cd /ibm/gpfs1/remotefset2/.snapshots/
[root@sonas3isv.mgmt001st001 .snapshots]# ls
snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-53
[root@sonas3isv.mgmt001st001 .snapshots]# cd snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-
53/

[root@sonas3isv.mgmt001st001 snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-53]# ls -lrt
total 0
-rw------- 1 root root 5242880000 Dec 16 03:13 sample1.mov

6. Add a new file, Sample5.mov to remotefset2 and remove the earlier sample1.mov file.
[root@sonas3isv.mgmt001st001 remotefset2]# ls
sample1.mov sample5.mov

[root@sonas3isv.mgmt001st001 remotefset2]# rm sample1.mov
[root@sonas3isv.mgmt001st001 remotefset2]# ls
sample5.mov

7. Take another peer snapshot, snap2.
[root@sonas3isv.mgmt001st001 /]# mkpsnap gpfs1 remotefset2 snap2
If the maximum limit of psnaps is reached, this will automatically delete the oldest snapshot.
Do you really want to perform the operation (yes/no - default no):yes
EFSSG0019I The psnap snap2 has been successfully created.

8. List the snapshots created in the previous steps.

[root@sonas3isv.mgmt001st001 /]# lspsnap gpfs1
Filesystem name Fileset name Snapshot ID Status Creation ID Timestamp

9. Verify the contents of snap2. Note that sample1.mov is deleted from snapshot2 and sample5.mov
is added to snap2
[root@sonas3isv.mgmt001st001 /]# cd /ibm/gpfs1/remotefset2/.snapshots/
snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-53 snap2-psnap-12402779243272901319-
AC1F8405:4EDFB110-12-11-12-16-03-29-42
[root@sonas3isv.mgmt001st001 .snapshots]# cd snap2-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-29-
42/

[root@sonas3isv.mgmt001st001 snap2-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-29-42]# ls

24

sample5.mov

At the home site:
Corresponding peer snapshots are automatically created on the home side (which reflect data
consistency)

1. List the created snapshots on the gpfs1 file system.
[root@sonas1isv.mgmt001st001]# lssnapshot gpfs1 -r
Device name Fileset name Snapshot ID Rule Name Status Creation Used (metadata)
Used (data) ID Timestamp
gpfs1 snap2-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-29-42 N/A Valid 12/16/11 3:29
AM 0 0 2 12/16/11 3:30 AM
gpfs1 snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-53 N/A Valid 12/16/11 3:22
AM 0 0 1 12/16/11 3:30 AM

2. Further explore the snapshot directory and compare the contents of the corresponding snapshots
with the contents of the cache site.
[root@sonas1isv.mgmt001st001 /]# cd /ibm/gpfs1/.snapshots/


[root@sonas1isv.mgmt001st001 .snapshots]# ls snap1-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-22-
53/cloud_nfs_src2
sample1.mov

[root@sonas1isv.mgmt001st001 .snapshots]# ls snap2-psnap-12402779243272901319-AC1F8405:4EDFB110-12-11-12-16-03-29-
42/cloud_nfs_src2
sample5.mov

Restoration of data at the home site
There can be instances when SONAS users might want to restore objects that are deleted accidentally or
in some cases want to restore previous version of the objects. Restores are performed only by request
and normally for small set of objects within a file system. As individual restore of objects within a file
system is allowed, restore commands also support wild cards. As opposed to the backup, where multiple
interface nodes might be configured to take backup on the Tivoli Storage Manager server, restore can only
be performed on the single interface node.
The data can be restored using Tivoli Storage Manager again at the home site. In this example, after
restoring the original file, the changes that were done at the remote site are lost.
You can use the startrestore command to restore SONAS data from the Tivoli Storage Manager server.
This command not only specifies the file pattern and the information on where the file system is mounted
but also the location where it has to be restored. Users cannot restore multiple file systems at the same
time.

25

Perform the following steps to restore data at home site.
1. Restore the file at a different location by using the -T option at the /ibm/gpfs1/test_restore
location.

[root@sonas1isv.mgmt001st001 ~]# startrestore '/ibm/gpfs1/cloud_nfs_src1/clip-2011-12-21 14;16;39.mov' -T
/ibm/gpfs1/test_restore/
EFSSA0184I The restore is started on gpfs1 with JobID 5.

2. Verify command completion and then check the restore log.
[root@sonas1isv.mgmt001st001 ~]# showlog 5
Job ID : 5
2012-01-10 23:17:31+07:00 Restorepattern: /ibm/gpfs1/cloud_nfs_src1/clip-2011-12-21 14;16;39.mov
2012-01-10 23:17:31+07:00 IBM Tivoli Storage Manager
2012-01-10 23:17:31+07:00 Command Line Backup-Archive Client Interface
2012-01-10 23:17:31+07:00 Client Version 6, Release 3, Level 0.4
2012-01-10 23:17:31+07:00 Client date/time: 2012-01-10 23:17:31
2012-01-10 23:17:31+07:00 (c) Copyright by IBM Corporation and other(s) 1990, 2011. All Rights Reserved.
2012-01-10 23:17:31+07:00
2012-01-10 23:17:31+07:00 Node Name: SONAS1INT04
2012-01-10 23:17:32+07:00 Session established with server ISVP13_LPM: AIX
2012-01-10 23:17:32+07:00 Server Version 6, Release 2, Level 2.2
2012-01-10 23:17:32+07:00 Server date/time: 2012-01-10 23:17:31 Last access: 2012-01-10 23:17:30
2012-01-10 23:17:32+07:00
2012-01-10 23:17:32+07:00 Accessing as node: SONAS1
2012-01-10 23:17:32+07:00 Restore function invoked.
2012-01-10 23:17:32+07:00
2012-01-10 23:17:35+07:00
2012-01-10 23:17:35+07:00 Restore processing finished.
2012-01-10 23:17:35+07:00
2012-01-10 23:17:35+07:00 Total number of objects restored: 2
2012-01-10 23:17:35+07:00 Total number of objects failed: 0
2012-01-10 23:17:35+07:00 Total number of bytes transferred: 15.00 MB
2012-01-10 23:17:35+07:00 Data transfer time: 0.12 sec
2012-01-10 23:17:35+07:00 Network data transfer rate: 125,461.99 KB/sec
2012-01-10 23:17:35+07:00 Aggregate data transfer rate: 4,951.26 KB/sec
2012-01-10 23:17:35+07:00 Elapsed processing time: 00:00:03
2012-01-10 23:17:35+07:00 dsmc return code: 0
----------------------------------------------------------
End of log - restore completed
----------------------------------------------------------


3. Verify that no errors are reported and then check the error log.
[root@sonas1isv.mgmt001st001 ~]# showerrors 5
Job ID : 5
----------------------------------------------------------
End of error - restore completed
----------------------------------------------------------


4. Verify the file after restoration.
[root@sonas1isv.mgmt001st001 test_restore]# ll
total 8
drwxrwxrwx 2 root root 8192 Dec 21 02:15 cloud_nfs_src1
[root@sonas1isv.mgmt001st001 test_restore]# cd cloud_nfs_src1/
[root@sonas1isv.mgmt001st001 cloud_nfs_src1]# ll

26

total 15368
-rw------- 1 root root 15734851 Dec 21 02:15 clip-2011-12-21 14;16;39.mov

For more detailed information on how to backup and restore files using TSM on SONAS, refer to the
references at the end of paper.

Two-way replication (home to remote and back )
This section refers to the movement of data from the home to the remote site and back from remote to the
home site.

From home site to cache site
Refer to the earlier section on “Initial data migration to cache site using Active Cloud Engine pre-
population and policy engine” for details.

From Cache site to home site
In continuation from the “Initial data migration to cache site using Active Cloud Engine pre-population
and policy engine” section, the next steps to replicate the data back to the home site in a different
(new) file set are as follows:

At the home site:
1. Create a new fileset, home_src2.
[root@sonas1isv.mgmt001st001 /]# mkfset gpfs1 home_src2
EFSSG0070I File set home_src2 created successfully.

2. Link the file set to the /ibm/gpfs1/home_src2 path.
[root@sonas1isv.mgmt001st001 /]# linkfset gpfs1 home_src2
EFSSG0078I File set home_src2 successfully linked.

3. Create a WAN cache source in the rw mode.
[root@sonas1isv.mgmt001st001 /]# mkwcachesource homeshare2 /ibm/gpfs1/home_src2 --client '9.11.82.90(rw)'

4. List the WAN caches and check whether the home WAN export is created.

[root@sonas1isv.mgmt001st001 /]# lswcachesource
WCache-Source Name WCache-Source Path ClientClusterId ClientClusterName WCache-Source Access Mode Is
Cached
homeshare1 "/ibm/gpfs1/home_src1" 12402779243272901319 sonas3isv.storage.tucson.ibm.com ro yes
homeshare2 /ibm/gpfs1/home_src2 12402779243272901319 sonas3isv.storage.tucson.ibm.com rw no

5. List the file sets.
[root@sonas1isv.mgmt001st001 home_src1]# lsfset gpfs1

27

ID Name Status Path Is independent Creation time Comment Timestamp
0 root Linked /ibm/gpfs1 yes 12/12/11 4:09 AM root fileset 1/11/12 12:06 AM

6. Verify that there are no contents at home site, before replication from cache site
[root@sonas1isv.mgmt001st001 /]# ll /ibm/gpfs1/home_src2
total 0

7. After replication from remotefset2, verify that the cache site is now populated with the replicated
data.
[root@sonas1isv.mgmt001st001 /]# ll /ibm/gpfs1/home_src2
total 15120
-rwx------ 1 root root 15734851 Jan 11 00:12 clip-2011-12-21 14;16;39.mov
drwx------ 3 root root 8192 Jan 11 00:12 Final Cut Events

At the remote site:
1. List WAN cache.

[root@sonas3isv.mgmt001st001 ~]# lswcache gpfs1
ID Name Status Path CreationTime Comment RemoteFilesetPath CacheState CacheMode
18 remotefset1 Linked /ibm/gpfs1/remotefset1 1/10/12 1:23 AM 9.11.82.28:/ibm/gpfs1/home_src1 enabled read-only

2. Create a new export in the rw mode, so that the Apple Final Cut Pro application can add this as a
SAN Location, (where the future read and write operation on the files will take place), when the
home or the primary site is down for various reasons.
[root@sonas3isv.mgmt001st001 remotefset1]# mkexport remoteshare1 /ibm/gpfs1/remotefset1 --nfs '*(rw,no_root_squash,insecure)'
EFSSG0019I The export remoteshare1 has been successfully created.

3. List the export.
[root@sonas3isv.mgmt001st001 remotefset1]# lsexport
Name Path Protocol Active Timestamp
remoteshare1 /ibm/gpfs1/remotefset1 NFS true 1/11/12 12:02 AM

4. Create a single-writer Active Cloud Engine file set.

[root@sonas3isv.mgmt001st001 /]# mkwcache gpfs1 remotefset2 /ibm/gpfs1/remotefset2 --cachemode single-writer --homeip
9.11.82.16 --remotepath /ibm/gpfs1/home_src2

5. Copy the initial replicated data from the remotest directory (where the data was initially replicated
using the prepopulation feature) to the newly created remotefeset2 file set. As the new caching
relationship is established between the cache and the home sites in sw mode, the newly copied
data (remotefset2) is automatically replicated back to home site in the newly created home_src2
file set.
[root@sonas3isv.mgmt001st001 /]# cp -R /ibm/gpfs1/remotefset1/* /ibm/gpfs1/remotefset2

28

Known limitation: When trying to replicate back from remote or cache site to home, first the data
needs to be placed in a new file set at the cache site. Then there are two options for home site:
 If the existing file set is deleted, then the replicated data can be transferred to same file
set (after creation)
 The replicated data can be transferred to a new file set, as an old file set already exists.
(Similar scenario is already explained in this paper.)

29

Summary
IBM SONAS system is a true and scalable solution for today's enterprise, file-based workloads and can be
fully integrated with the Tivoli Storage Manager server. Tivoli Backup-Archive client comes as a
preinstalled software with SONAS and data is reliably backed up directly to the external Tivoli Storage
Manager servers using interface nodes, providing faster performance. Active Cloud Engine on the other
hand helps to make sure that the Enterprise’s data is always replicated and updated across the data
centers spread worldwide. The prepopulation feature of Active Cloud Engine helps to transfer the initial
data from the home to the remote site. Also, the peer snapshots feature helps to ensure that data-
consistent snapshots are taken periodically across the home and remote sites, in case required for
restoration.
This paper provides the guidelines for backing up data on SONAS using Tivoli Storage Manager and
exploits the features of Active Cloud Engine in an Enterprise’s private cloud environment.

30

Acknowledgement
This paper could not have been completed without the valuable suggestions from Shekhar Agrawal, Active
Cloud Engine Team and Mandar Vaidya, ISV Team. The testing efforts went smooth with their deep
hands-on experience with Active Cloud Engine on SONAS and Tivoli Storage Manager. Author also likes
to thank Pratap Banthia and Kedar M Karmarkar for supporting this effort.

31

Appendix A: Resources
The following websites provide useful references to supplement the information contained in this paper:
 IBM SONAS administration and user documentation
http://publib.boulder.ibm.com/infocenter/sonasic/sonas1ic/index.jsp

 IBM Scale Out Network Attached Storage Administrator's Guide (GA32-0713)
http:/publib.boulder.ibm.com/infocenter/sonasic/sonas1ic/topic/com.ibm.sonas.doc/sonas_
admin_guide.pdf

 IBM SONAS Introduction and Planning Guide (GA32-0716)
http://publib.boulder.ibm.com/infocenter/sonasic/sonas1ic/topic/com.ibm.sonas.doc/sonas
_ipg.pdf

 IBM Scale Out Network Attached Storage: Architecture, Planning, and Implementation
Basics
ibm.com/redbooks/redbooks/pdfs/sg247875.pdf

 Backup and implementing IBM Tivoli Storage Manager solution on the IBM Storwize
V7000 Unified system
ibm.com/partnerworld/wps/servlet/ContentHandler/stg_ast_sto_wp_sonas_tivoli_storwize
_v700_unified_system

 Integrating IBM SONAS with IBM Tivoli Storage Manager
ibm.com/partnerworld/wps/servlet/ContentHandler/whitepaper/sonas/tivoli/use

32

Appendix B: Glossary
Application nodes – Nodes (such as interface nodes) within the cache cluster that handle application
requests where NFS server performs file operations to access data in order to serve applications on the
NFS client. By default, all the nodes that access a caching file set are application nodes.

Application request - A file system data or metadata request made by an application running on an
application node.
Cache cluster – A SONAS cluster where data is cached from the home cluster. It is also known as the
SONAS Active Cloud Engine cluster.
Cache server – A single server within the cache cluster.
Caching – A process where data from home SONAS cluster is brought into the caching SONAS cluster
upon access. The cached data is persistent and stored on the local SONAS cluster storage.
Caching file set – A GPFS file set that caches data from a single home cluster. GPFS can support
multiple caching file sets. A file set must be cache-enabled when it is created. Existing file sets, also
known as GPFS 3.3 file sets or file sets created before this feature is enabled, cannot be set for WAN
caching. Caching file sets are GPFS file sets and all the limitations to a normal file sets also apply to them.
Cache prepopulation daemon – A component of the prepopulation architecture that runs at the cache
cluster.
Disconnected mode – When a SONAS cache cluster loses connectivity to home cluster, it goes into a
state termed as disconnected mode.
FileRefreshLookupInterval, DirRefreshLookupInterval – After a file is cached in the cache cluster, data
is served from the cache itself. However, there is a need to perform a validation with the home cluster to
ensure that the cached file contents are up-to-date with the home cluster. For performance reasons,
revalidation is not performed on every file access, instead it is done when access happens after an interval
is passed. This interval is called refreshLookupInterval.
Gateway nodes – In Active Cloud Engine terms, these are nodes within the cache cluster that are
enabled for access to home cluster. These nodes use internal NFS clients to read and write data from the
home cluster. Active Cloud Engine Gateway nodes handle remote access for all Active Cloud Engine-
enabled file sets. In the SONAS environment, Active Cloud Engine gateway nodes are called as wcache
nodes.
GPFS – General Parallel File System and IBM cluster file system product.
Home cluster – The SONAS cluster where data to be pulled in, resides. The data is in the form of
exported file system through NFSv4 or parallel NFS. It is also known as the remote storage cluster.
Home server – A single interface node within the home cluster.
NAS cluster – NAS refers to an appliance or a device, which stores data and serves it over the network.
Today's NAS appliances includes multiple nodes/machines clustered together to make a NAS appliance.
Such appliances are referred to as NAS clusters in this paper.

NFS - A generic term referring to all the versions of the Network File System protocol.

33

Portable Operating System Interface (POSIX) –. All the file-related OS interfaces such as open, read,
write, close and so on are referred to as POSIX operations.

Prepopulation – The activity of bringing data into the cache cluster before it is accessed from the home
cluster .
Recovery mode - The mode of the Active Cloud Engine cluster when it is recovering from a failure
(gateway node, GPFS internal failures, such as queue drops, memory allocation failures, and so on). All
application requests are blocked in this mode.
Replication – The process of copying data or files from one SONAS cluster to another SONAS cluster.
There are many ways (such as rsync, explicit copy, and so on) to replicate data. Caching is another way to
achieve replication.
Revalidation – When the data is brought into the cache, it is necessary to ensure that it is up-to-date. The
process of verifying the validity of data with the home cluster is termed as revalidation. This normally
involves a NFS request to get the file attributes, (which contains the time when the file was changed or
modified at the home cluster) and comparing it with attributes stored from previous verification.

Wcache node – When Active Cloud Engine Gateway node is mapped into SONAS concepts, it is termed
as wcache node.
Synchronization lag - The time delay between the read at the home cluster and the time reflecting the
last write at the cache cluster. This attribute can be set using the mmchconfig command.
Validity lag - The time delay between the read at the cache cluster and the time reflecting the last write at
the home cluster. This attribute can be set system-wide or per file set using the
mmchfileset/mmchconfig command.
WAN – Wide area network connecting two SONAS clusters separated by locations.

34

About the authors
Gaurav Chhaunker, certified Project Management Professional (PMP), is a Senior Staff Software Engineer
in the IBM SONAS ISV Enablement Group. He has more than 8 years of experience working with various
Storage and System Management Technologies. Gaurav holds a Bachelor degree in Engineering from the
Kakatiya University, Andhra Pradesh, India and a Master degree in Computer and Information Science
from Cleveland State University, CSU, Ohio, USA. You can reach Gaurav at
gaurav.chhaunker@in.ibm.com.

35

Trademarks and special notices
© Copyright IBM Corporation 2012.
References in this document to IBM products or services do not imply that IBM intends to make them
available in every country.

IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business
Machines Corporation in the United States, other countries, or both. If these and other IBM trademarked
terms are marked on their first occurrence in this information with a trademark symbol (® or ™), these
symbols indicate U.S. registered or common law trademarks owned by IBM at the time this information
was published. Such trademarks may also be registered or common law trademarks in other countries. A
current list of IBM trademarks is available on the Web at "Copyright and trademark information" at
www.ibm.com/legal/copytrade.shtml.
Java and all Java-based trademarks and logos are trademarks or registered trademarks of Oracle and/or
its affiliates.
Microsoft, Windows, Windows NT, and the Windows logo are trademarks of Microsoft Corporation in the
United States, other countries, or both.
Intel, Intel Inside (logos), MMX, and Pentium are trademarks of Intel Corporation in the United States,
other countries, or both.
UNIX is a registered trademark of The Open Group in the United States and other countries.
Linux is a trademark of Linus Torvalds in the United States, other countries, or both.

Other company, product, or service names may be trademarks or service marks of others.
Information is provided "AS IS" without warranty of any kind.
All customer examples described are presented as illustrations of how those customers have used IBM
products and the results they may have achieved. Actual environmental costs and performance
characteristics may vary by customer.
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 and performance numbers are taken from publicly
available information, including vendor announcements and vendor worldwide homepages. IBM has not
tested these products and cannot confirm the accuracy of performance, capability, or any other claims
related to non-IBM products. 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. Contact your local IBM office or IBM authorized reseller for the
full text of the specific Statement of Direction.
Some information addresses anticipated future capabilities. Such information is not intended as a definitive
statement of a commitment to specific levels of performance, function or delivery schedules with respect to
any future products. Such commitments are only made in IBM product announcements. The information is

36

presented here to communicate IBM's current investment and development activities as a good faith effort
to help with our customers' future planning.

Performance is based on measurements and projections using standard IBM benchmarks in a controlled
environment. The actual throughput or 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.
Photographs shown are of engineering prototypes. Changes may be incorporated in production models.
Any references in this information to non-IBM websites are provided for convenience only and do not in
any manner serve as an endorsement of those websites. The materials at those websites are not part of
the materials for this IBM product and use of those websites is at your own risk.

37

IBM SONAS Enterprise backup and remote replication solution in a private cloud

More Related Content

What's hot (16)

Similar to IBM SONAS Enterprise backup and remote replication solution in a private cloud (20)

More from IBM India Smarter Computing (20)

Recently uploaded (20)

IBM SONAS Enterprise backup and remote replication solution in a private cloud