null  null
Storage Efficient Database Copy Options With
M ARCH 2015
Cloning a production database is an often needed procedure in customer environments to develop and
test new applications or test changes to an existing application. When a new application or additional
functionality is added, it calls for thorough testing using production data. Today this is achieved by
making a copy of the production data files in a test environment. In addition to the test environment,
copies of the production database are also made in one or more development environments where
application developers are creating their new applications and testing them. Each developer usually
wants to maintain his/her own copy of the production data. All of these potentially require a huge
amount of storage space to be allocated and managed.
When a customer addresses the above requirements by creating multiple environments with full size
production copies, they are faced with the following challenges:
(1) Each of these environments occupy the same amount of space as their parent database
(2) Creation of each environment consumes a lot of resources as the entire source database has to be
read and written
(3) It takes longer to create a full copy
What is needed is a different approach to creating multiple read/write copies of the same database
such that each copy can maintain their own changes without affecting other DBs and that does not
suffer the above challenges. We will refer to this approach as creating a storage efficient database
This ability to create storage efficient database copies is very useful to organizations. Many times
different members of a team need to perform isolated changes to their own copy of a database without
impacting others. In other words, there can be multiple DEV or TEST projects, each of them
performing updates to the base database, requiring isolated environments.
Exadata customers have various methods of creating storage efficient database copies. These
capabilities can be divided between:
» Capabilities that natively exist in Exadata
» Capabilities that can be used with Exadata but require additional hardware and/or licensing
Natively available capabilities in Exadata to create storage efficient database
The following options are natively available to Exadata customers to create storage efficient database copies.
(1) SPARSE disk group based database snapshots
SPARSE disk group based database snapshots is functionality included in Exadata Storage Software. It
requires Exadata Storage Software version and Oracle Database 12c version with
bundle patch 5 or later. This feature is designed to work on native Exadata ASM storage disk groups. It
uses ASM SPARSE grid disk based thin provisioning where snapshot databases created on a SPARSE
disk group need only the space for changes plus some metadata, thereby enabling storage efficient
snapshot databases.
Snapshot technology as deployed on Exadata is "allocate on first write", not copy on write. Snapshots start
with a shared read-only copy of the production database (or pluggable database (PDB)) that has been
cleansed of any sensitive information. When any of the child snapshots issues a write, the write is issued
to a private copy of that block inside that snapshot, preserving the base blocks for potential reads from
other snapshots. If, for the child database, this is a write of a block it has never written, then a new block is
allocated. Subsequent writes of the same block by the child will use the same, privately allocated, block.
Thus the allocation only happens on first write - thus the term "allocate on first write".
Multiple users can create independent snapshots from the same base database, therefore multiple test
and development environments can share space while maintaining independent databases for each task.
The base database must remain read-only during the usable lifespan of snapshots that are created from
that base database. If there is a need to create another set of snapshots from a more recent copy of
production data, a new read-only base from a production database needs to be created. If there are space
constraints for creating a new read-only base then existing snapshots can be dropped and the current
base database can be refreshed from production. Exadata database snapshots are integrated with the
Multitenant Database option to provide a simple interface for creating new PDB snapshots.
SPARSE disk group based snapshots on Exadata Storage Servers support all Oracle Exadata Storage
Server Software features such as Smart Scan, storage indexes and IORM, as the snapshots are stored in
ASM storage that is native to Exadata. This approach is recommended for Exadata customers as it is
available natively, fully integrated with ASM, and customers will benefit from the future enhancements
made to this functionality.
(2) ACFS Snapshots
Oracle ASM Cluster file system (ACFS) is a general-purpose POSIX, X/OPEN and Windows compliant file
system designed for single node and cluster configurations. An Oracle ACFS snapshot is an online, pointin-time copy of an Oracle ACFS file system. ACFS snapshots based on Copy on Write capability are
available on the Exadata platform when using Grid Infrastructure or later.
Oracle ACFS supports all database files and general purpose files on Oracle Exadata Database Machine
running Oracle Linux on database servers. Though the primary purpose of ACFS is to store general
purpose files, it also supports storing database files. On Exadata Database Machine, Oracle ACFS
supports the following database versions: Oracle Database 10g Rel. 2 ( and, Oracle
Database 11g ( and higher) and Oracle Database 12c ( and higher). In addition, all Oracle
ACFS advanced functionality is supported on Exadata. This includes: snapshots, replication, tagging,
security, encryption, audit and HANFS.
2 | Storage efficient Database copy options with exadata
Oracle ACFS is managed using native operating system commands as well as the Oracle ACFS
command line tool, acfsutil. Oracle ACFS is also fully integrated with Oracle Enterprise Manager Cloud
Control. Database files stored in ACFS on Exadata storage are subject to the following guidelines and
» Oracle ACFS replication or security/encryption/audit is only supported with general purpose files. These
functions are currently not supported with database files in an Oracle ACFS file system.
» ACFS snapshots should not be used to snapshot the live, running prod database on Exadata for
performance reasons.
» Oracle ACFS does not support the Exadata offload features (e.g. Smart Scan, Storage Indexes, IORM,
Network RM).
» Exadata Smart Flash Cache will cache read operations. Caching of write operations is expected in a later
» No specialized cache hints are passed from the database to the Exadata Storage layer, which means the
Smart Flash Cache heuristics are based on I/O size, similar to any other block storage caching
» Exadata Smart Flash Logging is not supported.
» Hardware Assisted Resilient Data (HARD) checks are not performed.
Please refer to Oracle ACFS Support on Oracle Exadata Database Machine (Linux only) (Doc ID
1929629.1) for additional information.
This is the only approach available on Exadata that supports Oracle Database 10g.
Capabilities that can be used with Exadata but require additional hardware and/or
These capabilities for creating storage efficient database copies with Exadata require Exadata to be
connected to an external storage system and are directly or indirectly based on the underlying storage
system’s Copy on Write (CoW) features. CoW operates on volumes (block devices), but have no native
understanding of the database topology.
Copy on Write (CoW) can be conceptually described as following:
“Copy-on-write stems from the understanding that when multiple separate tasks use initially identical
copies of some information (i.e., data stored in computer memory or disk storage), treating it as local data
that they may occasionally need to modify, then it is not necessary to immediately create separate copies
of that information for each task. Instead they can all be given pointers to the same resource, with the
provision that on the first occasion where they need to modify the data, they must first create a local copy
on which to perform the modification (the original resource remains unchanged). When there are many
separate processes all using the same resource, each with a small likelihood of having to modify it at all,
then it is possible to achieve significant resource savings by sharing resources this way.”
Since CoW capability depends on external storage, storage efficient databases created using this
technology on Exadata do not support Oracle Exadata Storage Server Software features such as Smart
Scan, storage indexes and IORM.
3 | Storage efficient Database copy options with exadata
The following options are available to Exadata customers who already have acquired or have plans to
acquire additional infrastructure to connect with Exadata Database Machine.
(1) Thin provisioning of PDBs using “Snapshot Copy” feature of Oracle 12c Multitenant Option when
Exadata is connected to ZFS Storage Appliance or NetApp:
ZFS Storage Appliance and NetApp are hybrid storage arrays designed to meet enterprise storage
needs. Oracle takes advantage of CoW capabilities of ZFS Storage Appliance and NetApp to provide
thin provisioning of PDBs using the “Snapshot Copy” feature of Oracle 12c multitenant option. In the
following example, the snapshot copy clause causes the connected appliances to perform thin
provisioning of the DEV44 PDB.
create pluggable database DEV44 from CRM_MASTER snapshot copy;
Instantaneous Database Clones with Oracle Multitenant explains specific details on configuring the
ZFS STORAGE APPLIANCE or NetApp appliance for snapshot cloning.
(2) Direct NFS Client CloneDB feature can be used for creating storage efficient copies of Exadata
database if the database backups are stored on an NFS share mounted on Exadata compute nodes.
The Direct NFS Client (dNFS) CloneDB feature utilizes CoW technology and was introduced in
Oracle Database 11g Release 1.dNFS is an implementation of the NFS client that runs as part of the
Oracle Database engine. Through this integration, Oracle Database optimizes the I/O access path to
the NFS server to provide improved performance, scalability, and reliability. By tuning the protocol to
match typical database I/O, dNFS provides faster performance than can be provided by the operating
system's NFS driver.
The CloneDB feature makes it possible to instantly clone an existing backup of a database mounted
over dNFS. The clone process uses Copy on Write technology; so only the changed data file blocks
need to be stored separately; the unmodified data is referenced directly from the backup files.
Exadata’s native backup solution is to back up Oracle databases to the Fast Recovery Area (FRA).
When Exadata is connected to network attached storage, the customer can choose to back up Oracle
databases to an NFS share mounted on the compute node. In this scenario, the dNFS CloneDB
feature can be used to create storage efficient copies of Oracle databases. Use of this feature will
also mean that database clones will be using non-ASM file storage.
For more information, see “How to Accelerate Test and Development through Rapid Cloning of
Production Databases and Operating Environments” white paper on CloneDB
(3) Enterprise Manager Snap Clone functionality is built on top of the Enterprise Manager 12c Storage
Management Framework (SMF) plug-in. The SMF plug-in provides a layer of abstraction to shield
DBAs and users from the nuances of the different storage systems. At the storage level, Snap Clone
accesses underlying storage technologies, such as Copy On Write or similar technologies, to perform
the required tasks. Snap Clone functionality can use either a hardware specific or hardware agnostic
4 | Storage efficient Database copy options with exadata
Hardware vendor specific solution requires customers to connect a storage subsystem with Exadata
and license clone and snapshot capabilities from the storage vendor.
This method, currently certified for ZFS Storage appliances and NetApp Storage appliances, enables
storage administrators to register storage appliances with Enterprise Manager, which then connects
directly to the storage appliance to perform all required snapshot and clone operations. A major
advantage of using ZFS Storage Appliance is the ability to support HCC seamlessly with Exadata.
No other vendor offers this. This approach requires you to license the relevant options on the storage
appliance, but it is the easiest, most efficient and fault tolerant approach.
Vendor agnostic software solution requires Solaris 11 file system (ZFS) on physical or virtual server
as a pre-requisite:
This method connects to the storage via an intermediate (ZFS) file system. This is a storage vendor
agnostic solution and can be used by any storage vendor, NAS or SAN. To configure this solution the
storage administrator mounts the volumes to a Solaris server and formats it with ZFS file system. The
ZFS file system then controls all snapshots and clone operations. The benefit of this approach is that
it does not require thin cloning options to be licensed on the storage since ZFS file system provides
these required capabilities. For additional details, check out the EM Snap Clone home page.
5 | Storage efficient Database copy options with exadata
by which
feature is
Core Technology
Additional Hardware
or licensing needed
Can cloned
DB support Minimum
Natively available capabilities in Exadata
group based
New type of ASM
Disk group called
group introduced in
Exadata S/W
Database File
support for
Exadata on ACFS
introduced in
Oracle Grid
Exadata Storage
Software version
Oracle RDBMS
12c and
Exadata Storage
Software Version
( is
recommended) and
Oracle Grid
or higher
Copy On Write of
the connected
storage subsystem
ZFS Storage Appliance
or NetApp Hardware
connected to Exadata.
License to use COW
technology of the
hardware vendor,
multitenant option of
Oracle RDBMS 12c
12c RDBMS with
multitenant option
12c and
implementation of
the NFS client that
runs as part of the
Oracle Database
NFS Share mounted
on Exadata.
11gR1 and
Copy On Write of
the connected
storage subsystem
Cloud Management
Pack for 12c, ZFS
Storage Appliance or
NetApp Hardware
connected to Exadata
if using the H/W
solution or Solaris 11
file system on a
physical or virtual
server connected to
Exadata if using a S/W
EM 12c + SMF
10g and
Capabilities that can be used with Exadata but requires additional hardware and/or licensing
provisioning of
PDBs using
Copy” feature
of Oracle 12c
Manager Snap
6 | Storage efficient Database copy options with exadata
Oracle Corporation, World Headquarters
Worldwide Inquiries
500 Oracle Parkway
Phone: +1.650.506.7000
Redwood Shores, CA 94065, USA
Fax: +1.650.506.7200
Copyright © 2015, Oracle and/or its affiliates. All rights reserved. This document is provided for information purposes only, and the
contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any other
warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or
fitness for a particular purpose. We specifically disclaim any liability with respect to this document, and no contractual obligations are
formed either directly or indirectly by this document. This document may not be reproduced or transmitted in any form or by any
means, electronic or mechanical, for any purpose, without our prior written permission.
Oracle and Java are registered trademarks of Oracle and/or its affiliates. Other names may be trademarks of their respective owners.
Intel and Intel Xeon are trademarks or registered trademarks of Intel Corporation. All SPARC trademarks are used under license and
are trademarks or registered trademarks of SPARC International, Inc. AMD, Opteron, the AMD logo, and the AMD Opteron logo are
trademarks or registered trademarks of Advanced Micro Devices. UNIX is a registered trademark of The Open Group.0115
March 2015
Author: Manish Shah
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF