| DISASTER
RECOVERY
JOURNAL
P. O. Box 510110
St. Louis, MO 63151
(314) 894-0276
Fax: (314) 894-7474
Internet
www.drj.com
E-mail drj@drj.com
EXECUTIVE PUBLISHER
Richard L. Arnold, CBCP
richard@drj.com
EDITOR-IN-CHIEF
Jon Seals
jon@drj.com
SENIOR
EDITOR
Janette Ballman
janette@drj.com
ASSOCIATE
EDITOR
Ed Pearce, CBCP
ed@drj.com
ASSISTANT EDITOR
Pamela Clifton
pamelaclifton@hotmail.com
COPY
EDITORS
Jim Hammill, CBCP
Richard Sandhofer
richards@drj.com
ADVERTISING
Robert Arnold
bob@drj.com
_____________
Corporate
President/CEO
Richard L. Arnold, CBCP
richard@drj.com
Vice
President
Robert Arnold
bob@drj.com
CONFERENCE COORDINATOR
Patti Fitzgerald, CBCP
patti@drj.com
CONFERENCE REGISTRAR
Merce Knese
mercedes@drj.com
CIRCULATION
Laura Baugh
laurab@drj.com
EXECUTIVE
COUNCIL
Mike Croy, Forsythe
Jeff Dato, MBCP, KPMG
John Jackson, IBM
Edward S. Devlin, E.S. Devlin & Associates
James Hammill, CBCP, JMH Consulting Inc.
Pat McAnally, SunGard Availability Services
Brian Turley, Strohl Systems
Belinda Wilson, Hewlett-Packard
INTERNATIONAL
CONTACTS
England: Thom Hetherington
Business Continuity
Phone: 0161-237-1007
thomh@tempus.demon.co.uk
Japan: Shinji Hosotsubo
Crisis Management and Preparedness Organization
Phone: 03-3519-6270
fax: 03-3519-6255
hosotsubo@cmpo.org
Brazil: José Carlos Ferreira
Disaster Recovery Mercosul
Phone and fax: 011-3666-9506
jocaff@uol.com.br
|
|
Click
Here for a Printable Version
SATA
RAID: Enterprise-Class Data Protection and Recovery for Everyone
By
STEVE MCINTOSH, CISSP
The greatest worry of system administrators in IT departments
used to be server and system downtime and the possibility of losing
data. Storage system threats such as viruses, malicious users, and disk
crashes were certainly high on this list. In recent years, however,
system administrators in the well-funded Fortune 500 IT departments
have worried less since they were able to afford RAID products. RAID,
or redundant arrays of independent disks, provides additional data integrity
and an I/O throughput boost for SCSI storage systems. This includes
enterprise-class RAID functionality such as “n-way mirroring”
(RAID1n), online raid-level migration (ORLM) and array “hiding”
that has typically been priced beyond the reach of individual workstation
and PC users as well as many IT departments in small- to medium-sized
businesses (SMBs).
One notable benefit of these RAID features is fast and easy backup/recovery
of local data arrays since they provide the ability to use the direct-attached
storage system to maintain an easily restored and secure backup of data.
This local RAID backup and restoration complements (but does not replace)
a company’s back-up procedures as it offers distinct benefits
not available with other back-up and restoration products.
For example, there is no need for special back-up hardware since the
backup is performed locally within the storage system. Also, data can
be restored quickly and easily since the system administrator does not
have to worry about precisely how to restore the data array from an
external backup device that resides somewhere out in the network or
even at a remote facility.
Wouldn’t it be wonderful if even individual workstation and PC
users could have such protection for their data, especially in light
of the recent blizzard of fast-moving viruses and e-mail worms?
For example, the Witty Worm is a recent plague and has spread like wildfire,
overwriting random sectors on the hard disk, preventing normal operation
of PCs and eventually causing PCs to crash. At the peak of the Witty
outbreak, according to an article in eWeek magazine, as many as 300,000
Witty-related packets per hour were recorded by some security services.
Another example is the recent MyDoom virus. This virus, which is another
speed demon that broke all previous speed-of-infection records, can
also destroy data on the hard drive.
Well, there is great news because, similar to the high-end, 3D graphic
functionality of the early 90s, premium RAID functionality is rapidly
finding its way to the mass market. As RAID products with enterprise-class
features converge with affordable interconnect and disk technologies
such as SATA (serial advanced technology attachment), the RAID features
utilized by the big IT departments are suddenly within the reach of
SMB IT staffs and even becoming available to individual workstation
and PC users. This means that many new opportunities suddenly emerge
for RAID’s ability to create secure local backups of data that
are protected from the likes of viruses, malicious users (or non-malicious
but very active toddlers), and disk crashes.
For example, RAID functionality can be used to support a software engineer
with sensitive product development data on his or her workstation, or
a CPA with clients’ tax forms on a home office PC, or a medical
practice administrator with patient data on the office server that should
be secured according to stringent HIPAA regulations.
Basic Requirements for RAID
The possibilities for deploying RAID technology abound. Wherever important
data is being stored, there is likely an opportunity for RAID. So how
does one start adding RAID functionality to a SMB or personal computing
environment? The first items required are storage components that support
SATA and enclosures that support these components. The good news is
that SATA equipment is a good deal more affordable than the SCSI gear
on which RAID traditionally operated. Items required include:
- A high-performance, highly functional SATA RAID controller;
- A minimum of three SATA disks;
- Enclosures that allow the configuration for at least four SATA
drives and provide adequate power and cooling; and
- A SATA hot-swap drive enclosure.
- good place to start getting guidance on these products is from
a local computer reseller who should be able to provide adequate configurations
for SATA-based RAID equipment.
Configuring RAID to Prepare for Disaster Recovery
To help understand how to utilize RAID for disaster recovery, let’s
first review some RAID basics. RAID allows the creation of a data array
that appears as a single volume or disk to the user but, under the covers,
the data actually resides on multiple disks to enhance data integrity
or I/O performance. There are different types of RAID arrays. With a
RAID1 or mirrored array, data integrity is enhanced since the same data
is written to two physical disks. If one disk fails, the data is still
on the other disk. “N-way mirroring” or RAID1n is an enhancement
of RAID1 that creates more than one mirror of existing data.
Mirroring is a basic RAID feature that is now available with some operating
systems such as Linux and Windows. However, to create a secure local
backup with RAID, the enterprise-class functionality mentioned earlier
must be deployed, including:
- RAID1n – Create more than one copy of a mirror;
- Mirror splitting – Split a mirror off from one array into
its own, new array;
- Array hiding – Hide an array backup from users and from the
operating system, making it visible only to the system administrator;
and
- Online RAID level migration (ORLM) – Transform an array from
one type to another without interruption to I/O processing.
1. Create a 3-way RAID1n array
The first step in creating a secure local backup is to make a 3-way
RAID1n array that essentially creates three copies of your data.
Figure 1. A 3-way RAID1n array

Once the array is created, the data is protected from a disk crash
or failure, but not from malicious users or viruses since all mirrors
are still visible and accessible to users and the operating system.
2. Split the mirror and hide an array
The most functional SATA RAID controllers support mirror splitting and
array hiding. These features treat one of the array mirrors as a data
backup that can be separated from the active array and then hidden from
users and the operating system. Once the mirror has been split and hidden,
its data is no longer susceptible to viruses or malicious users since
they cannot corrupt or delete what they cannot see.
Figure 2: A split mirror and hidden array

At this point, there is an active RAID1 array plus a hidden backup
of the data which is now protected from both disk crashes and from ill-behaved
software and users.
Disaster Recovery
Now, imagine that a catastrophe occurs: the anti-virus software does
not recognize a new, fast-spreading virus and it infects the system,
overwriting critical data in the active RAID1 array.
3. Unhide the hidden array and re-create
the RAID1 mirror
Having removed the virus, the damaged array is deleted. Then, using
ORLM, the hidden array is transformed to a visible RAID1 array. At this
point, the new array is split and re-hidden, and the system is back
up and running as before.
Figure 3: Disaster recovery

The real benefit of using local RAID is that systems can now recover
quickly and easily from a serious disk or data catastrophe without any
special backup hardware purchases.
However, there is a need to be diligent about periodically creating
and hiding an updated mirror because it contains the data at the point
when the split occurs, and any modification made to a visible, active
array subsequent to each split is not reflected in the hidden array.
Final Notes
There are a couple of notes worth mentioning before closing. First,
RAID is not a replacement for anti-virus software; it complements it.
Anti-virus products do an excellent job of screening known viruses.
However, should a new, unknown virus evade the AV screen and corrupt
data, recovery is possible with a hidden array solution. Second, RAID
complements rather than replaces any backup procedures that an organization
may already have in place. For example, a hidden, local array is protected
from software or user threats, but not from fires or floods. Thus, a
remote backup procedure is still required for business-critical data.
So, what are the “take-aways” from all this?
- Data is important and critical to most companies and businesses,
yet the threats are flourishing faster than disaster prevention products
can respond to.
- The RAID features that brought peace-of-mind to Fortune 500 companies
are no longer beyond the budgets of SMBs and consumers.
- By using the right RAID in the right way, all of us can begin to
sleep better at night like system admins who have long enjoyed the
benefits of RAID.
Steve McIntosh, MBA, CISSP, is a product marketing manager for Broadcom
Corporation. He also has more than 10 years in security software product
management and customer support at nCipher, CertCo, and Digital Equipment
Corporation (now HP).
©Copyright
2004 Systems Support Inc. All rights reserved. Reproduction in whole
or in part in any form or medium without the express written permission
of System Support Inc. is prohibited.
|