As it turned out, the
brilliance and competence of Amazon’s technocracy were no match for simple
human error. An incorrect manual update to the network set up a domino effect
of Catch-22-esque compound failures to Amazon’s EBS (Elastic Block Store)
details of which can be found in
a 6 page explanation the company offered last week.
Although cloud naysayers
will no doubt try to make this event the poster child for Luddite agendas, it
won’t work. Data was lost. Business was lost. News was made. But overall, the fallout
has not been too bad as users have quickly come forward to stand by Amazon and
the choice to use its services.
And the risk was largely knowable.
As Jason posted in an article On Cascading Failures and Amazon’s Elastic
Block Store “This is not a
“speed bump” or a “cloud failure” or “growing pains”, this is a foreseeable
consequence of fundamental architectural decisions made by Amazon.“
The gains surpassed the
risk. And will continue to do so. Though still in its early stages, life in the
cloud is really a lot more of a learning experience than it is an adventure. Everyone
involved will continue to get a little wiser in the ways that only experience –
usually bad ones – can confer on those gaining the wisdom.
As the only software
vendor offering patented technology to virtualize Windows, Linux, and Solaris
server applications – AppZero is a huge
fan of all things cloud. We invite our customers to move their applications to
and from datacenter and cloud(s) and cloud to cloud, with no lock-in. I’ve seen
the hesitancy that comes from skepticism and trepidation as well as the high
fives and smiles that accompany seeing and believing.
But here’s something I’ve
also seen that I will never understand: Organizations running very important
(though not “mission critical”) Solaris
2.6 applications on hardware that is past the hope of life support. The
hardware will fail and take with it the applications. When I say, “It’s not a
matter of if … it’s when.” I get knowing chuckles and head
shakes as IT pros tell me how very right I am.
When I go on to tell them
that AppZero software can encapsulate their Solaris 2.6 and 7 applications, pick
them up, and deposit them on Solaris 10 and bright shiny, inexpensive, reliable
machines, all without a line of code … they are intrigued. Of course they are. Here’s
a very cost-effective solution to a guaranteed problem.
Okay. So here’s the question: Then why doesn’t
every one of them just jump up and sign on with AppZero? Human nature or human
error? Have they lived so long in denial that they’ve crossed over into magical
thinking, convinced that because it hasn’t happened … it won’t? Or do they
expect to be in new jobs before the when = now?
Cloud risks pale against
cloud gain. Life on the San Andreas Fault brings a great lifestyle until the
big one. But important apps on Solaris 2.6?
I don’t have a clue so feel free to send me one. In the meantime, let’s cloudify
those Windows and Linux client/server apps – no code, no lock-in, no pain.
I am always looking for a
way to communicate better and cut to the heart of any discussion. So, if you
have thoughts on this subject, drop me a line at GregO {@} Appzero {dot} com or tweet me at http://twitter.com/gregoryjoconnor
del.icio.us
Digg
Facebook
LinkedIn
Reddit
SlashDot
StumbleUpon
Twitter
