Helder Almeida - Fotolia

Evaluate Weigh the pros and cons of technologies, products and projects you are considering.

Cloud outage stresses provider-user responsibilities

Cloud outages leave customers scrambling without access to apps and services, but providers aren't solely to blame. Provider-user teamwork is key to availability.

Cloud outages are an unfortunate and unavoidable part of the cloud. And while no one likes being without services for any period of time, a lack of communication was the thorn in Microsoft cloud customers' sides following the recent Azure outage.

"A common pattern in outages is that people don't complain about the outages, they complain about the fact that they were caught with their pants down, so to speak," said David Linthicum. "They just couldn't react to it, couldn't respond to it or get information out to their users."

Linthicum discussed the cloud outage and other topics with Janakiram MSV, head of cloud infrastructure at Aditi Technologies, a Microsoft partner. Topics of discussion include:

1. Is Microsoft completely to blame for its cloud outage? Janakiram and Linthicum agree that cloud management and monitoring is a team effort.

"The problem is customers don't realize that uptime and availability is a shared responsibility," Janakiram said. "One key thing is to understand and realize is that cloud will fail, cloud is not a silver bullet solution. You can’t blame Microsoft … because they are like any other infrastructure providers and are bound to go through these hurdles of downtime and availability issues."

Who is ultimately to blame?

"Bottom line, you can't just blame the cloud service provider," Janakiram added. "Of course, they need to deliver on their [service-level agreements], but having a very strong partner who can complement your skills and deliver the right value is very important for you to get the best value out of your cloud deployments."

Is it a cloud architecture issue? Customers should take a pessimistic approach to cloud deployments, architecture and design and ensure applications can survive availability issues, according to Janakiram. (4:00-18:28)

2. IBM continues to make its cloud market push, but what is its long-term prognosis? "IBM typically doesn't build stuff, they buy stuff … I don't think they're going to be able to acquire enough companies and integrate enough companies to keep up with the market space," Linthicum said.

IBM's strong enterprise relationship and credibility will help the vendor in the cloud market, Janakiram said. Where will IBM stand in the cloud space in five years?

"I just don't see a future out there for these guys unless they make some drastic changes internally and start shifting their technology to areas where it's going to be more innovative and more valuable to the enterprise," Linthicum added. (18:29-24:56)

Next Steps

Office 365 silent patch creates customer problems

Cloud's potential a concern after Azure outage

Cloud disaster recovery guide to plan ahead

Dig Deeper on Public cloud and other cloud deployment models

Join the conversation


Send me notifications when other members comment.

Please create a username to comment.

That is the Achille's heels of cloud applications, in that if your cloud provider takes a hit, you take a hit. Having a way to make sure that the provider can communicate the problem so that users/customers on the other end can react is hugely important. Customers can be understanding if they realize a situation is beyond anyone's control, but silence doesn't foster that.
Listen to 23:25 and forward. First time I heard refuted "No one gets fired for hiring/picking IBM" and the refute is "oh yes they have"... "that was the 80s buddy". Good to hear reality get in the mix vs the corporate CYA comment. :)