Amazon S3 outage: A guide to getting over cloud failures
A comprehensive collection of articles, videos and more, hand-picked by our editors
For many applications, an Amazon or Azure cloud might be overkill. In such cases, a managed services model approach just might be the better option, and here's why.
It can be a tad disconcerting when a popular trend pushes its way through the industry and you and your organization are yet to jump on the bandwagon. For enterprises that haven't yet moved their applications onto the Azure, Google or Amazon cloud, it would be understandable for managers and C-level executives to be questioning both why it hasn't happened yet and when it actually will. But according to Jordan Jacobs, vice president of products at SingleHop, the Azure, Amazon and Google cloud models are being oversold, and for many core business functions, a managed services approach to application hosting is often a better model.
Public clouds vs. managed services model
"The thing that wows a lot of people is the market share discrepancy between public clouds and managed services, especially when compared to the press each one gets," said Jacobs. "Amazon, Azure and Google get all of the press, but they're actually only about a third of the managed services and managed hosting market."
Unfortunately, the love affair the press is having with the dominant cloud providers is causing a great deal of consternation with decision makers. On the one hand, decision makers feel that they need to catch up with the latest trend; on the other hand, they are having a hard time rationalizing, in terms of cost efficiencies, security, and business benefits, the porting of their core business applications into the public cloud. It's creating a sort of analysis paralysis, where decision makers are unsure of whether using the public cloud is the right move, whether the managed services model makes more sense or if they should just keep everything on premises.
"Everybody is hearing buzzwords and they're worried that they're doing something wrong," said Jacobs. The constant talk about moving to the cloud, harnessing big data, deploying to Docker containers and architecting applications with microservices is making software professionals question the processes they currently have in place, even though those processes may, in fact, be working. "It's this funny place where everyone is worried that they're doing it wrong, but everyone is doing the same thing."
So what's the right approach to moving applications away from an on-premises model and taking advantage of economies of scale, while not getting overwhelmed by the multitude of options? For the applications that need cloud bursting and whose unpredictable usage spikes demand elasticity, the Azure, Amazon and Google options are likely the best fit. However, those boring applications, like expense tracking or email hosting, can garner significant cost savings when moved to a managed services model. Not only would a managed services model take advantage of economies of scale, but it would also offset the often more significant human cost of management and administration.
To hear more about the differences between the public cloud and a managed services model, and to hear about the Cloud Expo sessions entitled "Lost efficiencies of the public cloud" and "What cloud vendors don't want you to know," listen to the accompanying podcast with Jordan Jacobs and TheServerSide's Cameron McKenzie.
Research suggests MSPs struggle to find the optimal revenue mix
Why are MSPs still doing break/fix work?
Cybersecurity services: MSPs look to expand portfolios