In life, there are many things that we all take for granted. These
are everyday things that always just seem to be there and are only
conspicuous when they are absent. In the last week two of these items
have shown that they are not as enduring as many might have thought.
Google and Amazon have both had major outages to their services that
impacted internet traffic. These two mysterious outages have caused
some concern about the reliability of both companies as well as the
possibility of a larger issue (perhaps a potential attack). However,
neither Google nor Amazon are coming clean about these incidents making
many even more nervous.
The rash of outages kicked off on Saturday the 17th at about 4:47PM
(Pacific Time) with Google Apps disappearing from the internet. Although
the entire outage did not last more than about 5 minutes the effects
were pretty staggering. At the height of the incident worldwide internet
traffic dropped by an estimated 40%. Yes you read that right almost
half of the global internet traffic was affected by a five minute outage
at Google.
This number seems pretty large (and it is) however, we have a
feeling that the majority of that traffic was spam heading to Gmail.
Many companies that rely on Google were at a standstill for the unusual
outage with many companies at a loss for what to do or even who to
report it to. I know that I have already been contacted about
contingency plans for cloud service outages as a result of this brief,
unplanned down time for Google.
Now, while Google’s outage was shocking and left many stunned and
confused; Amazon’s outage appeared to be more “business as usual” even
though it was highly annoying to many Amazon users. Here the outage was
just the latest in a string of them over the last 12 months. Amazon’s
hiccup lasted a little longer than Google’s with the total outage time
around one hour (11:50 am to 12:44PM PST).
During the time of the outage the Amazon Web Services was
unavailable and even Amazon’s own site was offline (trace routes dropped
off en-route to their datacenter). Twitter was abuzz with comments and
complaints about the problem, but nothing ever showed up on the AWS
dashboard. There was no attempt to identify what percentage of internet
traffic the Amazon outage impacted which is interesting considering how
many were eager to see how much of an impact Google had. Our guess it
that Amazon has these issues far too often to be really big news
anymore.
Internet and service outages are nothing new and will continue, but
when you look at the impact they can have (as in the case of Google’s
outage) it is not hard to see just what a target companies like Google,
Amazon and Microsoft are. We are not talking about from the average
hacker or even Anonymous; we are talking about the much larger concern
of state sponsored attacks.
How many companies would be dead in the water if they could no
longer reach Google, Amazon, Microsoft or any number of cloud services?
Sure a 5 minute outage out of an entire year is very small (about
0.00095%), but it shows that even Google’s services are not
invulnerable.
As for Amazon, well there are still a large number of companies that
rely on their services that continue to be impacted by their frequent
outages, but many of these do not have the resources to bring all of
their systems back in house without an even larger impact to their
business. It makes me wonder just what these businesses were told to get
them to sign up in the first place.
No comments:
Post a Comment