Provider inside
Both infrastructure as a service (IaaS) and
platform as a service (PaaS) services are ideal for third party developers to
create service offerings or software as a service (SaaS) solutions ready for
customers. For instance NetFlix, LinkedIn, and Salesforce are the services
powered by PaaS, and IaaS core services but are delivered as a customer ready
SaaS service. As such a Netflix user
does not contemplate how Amazon AWS makes their IaaS service work to ensure
that they can watch a movie. Bottom line users watch their movies. This model
is really the future of cloud services.
However unlike the
telecommunications companies that deliver our phone services end to end, cloud
solutions like Netflix have a dependency model users should be aware of.
Ultimately users should have the right to know who is touching, and managing
their data. And providers should be required to expose the information.
Let's consider this.
The videos serviced
by Netflix powered by Amazon AWS requires that Amazon hosting, storage, name
resolution, and many others be available, and Netflix provisioning, account
management, payment, billing, and many more services be online. The
connectivity between the front end services and virtual machines in AWS must
run flawlessly, and overall provide a seamless service so that customers can
sit back and watch a movie, without complaining about the wait time of their
log in, or video playback. The result of
an outage in AWS had broad impact on their customers including Netflix ability
to provide their services. Clearly both Amazon and Netflix have worked hard to
make their solution redundant and failure resistant, but ultimately Amazon
faced the wrath of their direct customers (kindle users) as well as corporate
customers like Netflix. Try as they might Netflix was ultimately taken offline
by an Amazon service failure. http://techblog.netflix.com/2011/04/lessons-netflix-learned-from-aws-outage.html.
What is fascinating
is that at the end of the day, Netflix can blame Amazon and claim that they
have done everything they can to ensure quality of service. And Amazon can
dance around the fact that they tried everything to prevent their outage. The
truth is Netflix is right. Amazon failed. But what's misleading is that Netflix
should also be held at fault. For instance why did they not use multiple cloud
providers to reduce the chance of failure, and truly claim elastic computing?
And as a user I
should have been able to know how and what Amazon was doing to resolve my issue
with Netflix, right?
Comments
Post a Comment