Computers – Convert a Major Catastrophe into a Minor Inconvenience
How will you handle things when a critical computer, one
upon which you or your organisation relies, ceases to work? This could be
caused by something like a disk dying, by the machine being stolen, by fire or
flood. All of those things are totally
beyond your control and can happen at any time.
The risk is caused by there being a single point of failure,
which is very risky indeed!
“We have backups to handle this situation” is the normal reply. Well, that’s the bare minimum, a starting
point, for without backups you are deader than dead!
But, how often do you test that these backups will in fact
allow you attain a fully operational condition?
How long does it take to rebuild the situation?
What is the impact on your business? Is this impact
affordable?
All of the above are critical questions and need to be
answered fully and truthfully. Are you
aware, that as a director or partner, you can be held responsible for
maintaining good governance? Disaster Recovery is a measurable of Good
Governance (King II & King III).
Makes one think, doesn’t it?
Fortunately there are readily available options which will
allow for a fast recovery of the situation.
These are not new and have been in use by the major
corporate players for a long time. The drawback is that until now, the cost has
been extremely high, but this has changed – now there are options based upon
the same techniques and processes, but at a fraction of the cost.
The solution is based on something called a Virtual Machine.
Simply put, a Virtual Machine is software which processes a set of disk files
which describe a physical computer. The beauty of the solution is that when the
Virtual Machine is running, it looks and operates exactly like the physical
machine it is emulating! In fact there
is a major push to move physical computers to their virtual counterparts,
especially where an organisation has five or more machines it its computer
room, but that will be the agenda for a future article.
How does this help me?
Well, if a virtual copy can be made of each of your critical
computers, then these Virtual Machines can be used in place of the physical
one.
As a Virtual Machine is simply a set of data files, it can
be stored on an off site disk drive.
This removes the risk of a single point of failure.
The Virtual Machine can be up and running with very little
delay. What kind of delay can I expect?
This is where the rule “The more you spend, the faster the recovery” applies.
The large financial corporate institutions will spend millions to keep the
delay down to thousandths of a second, but a single laptop user can probably
tolerate being down for a few hours.
While recovery is a critical part of the process, so is
having the ability to test the process without risk. For instance, at the
beginning of this article I asked how often your backups were tested. It is
amazing how unreliable untested backups are, and, being in the middle of a
crisis is not the best time to discover that your backup is inadequate!
If you have a Virtual Machine it is a simple exercise to run
a full restore and then to fully test that it has accomplished what was
expected. This test is exactly the same as if it had taken place on the
original machine, but it can be undone and redone as many times as is necessary
in order to guarantee that it works 100% correctly.
In fact a full Disaster Recovery test should be carried out,
at least once a year, as a matter of policy.
