The advantages of virtualization are undeniable. There are tremendous operational efficiencies to be had in many different areas, but it's not a cure-all. Delivering high availability or disaster recovery to meet demands of modern 24/7 operation means understanding all your risk points and architecting around them. Virtualization won't deliver business continuity for free, but as Knowledge Center contributor Andrew Barnes explains, by architecting your virtual environment with the right tools and experience, you'll greatly ease the process and increase your chances for success-at a cost that's right for business.
one of the most disruptive technologies of recent times, virtualization
is driving change in IT shops large and small. Yet, does that change go
far enough? For as long as business has relied on IT, there has been an
underlying fear about the possible failure of underlying systems'
components and operational disruption.
Of course, the biggest concern has been loss of vital business data,
resulting in the evolution of today's sophisticated data recovery
solutions. Virtualization can add tremendous value in ensuring business
continuity, but not on its own. It's critical to look at the
combination of physical and virtual to determine the best overall
approach for your business.
Industry pundits talk a lot about focusing on recovery. "Recovery
management" or "recoverability" is the in-vogue term, yet this minor
switch of emphasis ignores one key fact: recovery
is a reaction to failure
By the time a recovery is complete, the damage has been done.
Productivity has been hit, communication has been disrupted and
business has been lost. No matter which way you look, in a business
world that gets more competitive by the day, critical systems need to
be continuously available. That means operations servicing customers,
partners and employees 24/7, no matter where in the world they are.
Building an IT strategy to deliver continuous availability
business-critical systems should go beyond using legacy backup and data
protection principles. Bringing together virtualization, automation,
and application monitoring and replication technologies to ensure that business continues
accelerates the value of virtualization far beyond the server room.
Address the real causes of downtime
Let's face it: although the prospect of a real
clearly exists, in all actuality, a facilities failure, application
software failure or IT failure is more likely to be the source of a
business disaster than a flood, hurricane, earthquake or terrorist
Take databases. One day of database downtime, caused by the failure
of a cooling unit in a server room, can spell disaster for a company in
terms of revenue, productivity and reputation. On another level,
application failure due to a poorly-timed patch, for instance, is a
much more likely threat to availability than a physical server outage.
Wherever a business-critical application is deployed, the threat of
downtime comes from many sources. Delivering the right infrastructure
to address these threats means the difference between ultimate success
Virtualization and application availability: B- at best
Most virtualization vendors' availability solutions start by
addressing entire host failover using platform high-availability
facilities. While these protect against a physical server failure, they
ignore the fact that the failure of an application within a virtual
machine is much more likely to bring business to a halt. Facilities to
detect application failure within a VM are nonexistent. This
shortcoming is further complicated by the fact that a "blue-screened"
VM may still appear active.
Another aspect that's not often discussed is the new and/or
refreshed infrastructure required to achieve high availability and
business continuity, which in today's economic climate may not be
possible from a budget perspective. Virtualization platform-based
failover relies on shared storage, which can be vulnerable to failures
as well. This approach also relies on the premise that the protected
applications are tolerant of running in a virtual world, yet many
administrators still have doubts about memory, CPU and I/O requirements
in a virtual environment.
So, while virtualization brings progress, the ability to deliver
continuous availability through virtualization alone is a ways off. If
you still rely on physical deployments, then additional strategies must