When preparing for a disaster, the ability to
fail over to a remote location with a minimum of downtime can
definitely offer advantages. While remote-availability solutions
offer both pros and cons, these solutions play a role in many
business continuity plans.
On the plus side, RA offers the ability to
resume data-system activities in another physical location in the
event that the primary location becomes unavailable. While you may
think that losing an entire physical location and/or data center to
a disaster sounds like an unlikely occurrence, recent history
offers several cases in point.
The blackout that plowed through the
northeastern United States last August is one example; the series
of devastating hurricanes in Florida earlier this year is another.
Both of these disasters were unexpected, offered little–if
any–opportunity for preparation or immediate action, and
completely cut off employees and external resources from internal
data systems. Those companies that had enacted RA solutions were
able to get their information systems up and running at another
physical location, one unaffected by either localized or more
Tornados, fires, floods, blackouts, and other
natural or even man-made disasters often give no warning, and
nearly all of them cause at least temporary outages of primary data
facilities. Without RA, once your uninterruptible power supply
(UPS) or diesel generators give out, you’re out of luck.
But RA systems also have some drawbacks. First
of all, they’re almost always a massive expense. You need space to
store the alternate systems, which can be expensive.
While it’s preferable to have this space
completely under your organization’s control, you can also rent or
share space with other companies. But this space typically comes at
a premium, since it traditionally offers data-center-level
architecture and infrastructure. Your company could cut costs by
using normal office space in another location, but this risks your
failover solution causing a failure itself.
In addition, you need to provide redundant
hardware. While it’s often possible to protect multiple systems
with a single system on the RA side, you must be aware of
applications that don’t support a many-to-one failover. For
example, single-instance applications (such as Microsoft Exchange
Server and many fax applications) nearly always require a
one-to-one configuration for failover, both locally and
You must also ensure that you have enough
bandwidth to the remote site to support replication of data via one
or many methods. Hardware and software replication tools both
require bandwidth, and different systems of the same general type
can widely vary in their requirements.
Your organization may also need to transport
backup media such as tapes and optical drive media to the DR site
in order to perform restoration of those systems and components
that you can’t replicate for one reason or another.
In spite of these concerns, RA solutions are a
very valid and, in many cases, vital part of an overall business
continuity plan. If your organization can withstand the budget and
other requirements, it’s a great way to ensure that your systems
can go beyond a single-site disaster.
How well can your organization deal with an emergency? Automatically sign up for our free Disaster Recovery newsletter, delivered each Tuesday, and make sure you’re prepared for the next catastrophe.