By Bob Gourley
We previously brought your attention to a Defense Science Board study on Cloud Computing (see: DSB Report on Reliability in a Digital Cloud )
I continue to believe this report captures critically important, actionable information and makes conclusions that should form the underpinning of DoD IT modernization efforts. It is a good read. The topic of the report was brought to mind again today because of the most excellent Wired article by Cade Metz titled “Why Some Startups Say the cloud is a waste ”
In the article, Metz reports that:
In Silicon Valley, tech startups typically build their businesses with help from cloud computing services — services that provide instant access to computing power via the internet — and Frenkiel’s startup, a San Francisco outfit called MemSQL, was no exception. It rented computing power from the granddaddy of cloud computing, Amazon.com.
But in May, about two years after MemSQL was founded, Frenkiel and company came down from the Amazon cloud, moving most of their operation onto a fleet of good old fashioned computers they could actually put their hands on. They had reached the point where physical machines were cheaper — much, much cheaper — than the virtual machines available from Amazon. “I’m not a big believer in the public cloud,” Frenkiel says. “It’s just not effective in the long run.”
Frenkiel’s story shows that while cloud computing is suited to many tasks — including getting your startup off the ground or running a modest website — it doesn’t make sense for others. When Zynga’s online gaming empire expanded to epic sizes in 2012, the company made headlines in shifting much of its operation off the Amazon cloud and into its own data centers, but smaller operations are making the move too.
Like MemSQL, the ride-sharing startup Uber recently moved most of its tech off the Amazon cloud, according to the company that now houses its physical servers, Peak Hosting. And various others, from analytics outfit MixPanel to online clothes-trading startup Tradesy, have disclosed similar shifts.
Now here is a bit from the CTOvision summary of the DSB report on cloud computing:
The task force offers important recommendations for DoD focused on identifying and applying cloud computing resources to DoD mission areas; improving DoD’s implementation of cloud computing, enhancing cloud resiliency in degraded operations and in identifying areas requiring further research and development.
The bottom line of this report: For sensitive data, well engineered private clouds are a better approach than any public cloud option at this point.
In the words of the task force co-chairs: ”The Task Force recommends that for sensitive, classified, or time-critical applications, the DoD should pursue private cloud computing to enhance mission capabilities, provided that strong security measures are in place. This report recommends several improvements in cloud computing implementations to strengthen cyber security and reliability.”
There is special focus on modern cloud computing architectures and data center designs supportive of mission needs. There is an entire section on the architecture of the modern cloud data center. This section lays out very well the benefits of modular systems and underscores their finding and recommendation on modular data centers.
To me the big point of both the DSB report and Cade Metz’s piece are that the right answer to Cloud Computing questions will always depend on the mission need of the organization, and that goes for organizations of all sizes. At the small end of the scale, I can’t imagine I would ever stop leveraging cloud computing capabilities for my firm’s email, document managment, web hosting, customer relationship management, prototyping and most other business functions. The cost is very low, the agility is excellent, and when done right the risks can be mitigated. It is also easy to end services (you can stop instantly). I think most businesses see the benefits. We do these things because we have a purpose and objective and a reason. At the higher end there will, of course, also be purposes and reasons and missions that cloud computing benefits can serve. But other times the mission will not be addressable by cloud computing solutions.