Increasingly, enterprises want the rapid provisioning and elasticity, self-service capabilities and on-demand advantages of cloud applications and services. But, they also want enterprise-class security, availability and manageability. Unfortunately, many are finding that’s easier said than done.
On the one hand, enterprises trying to build their own private clouds have found out the hard way that it’s a challenging undertaking. At the other end of the spectrum, public clouds lack the requirements needed for enterprise applications, especially security and availability. So it stands to reason that in this “Goldilocks” choice, the third option – turning to a managed cloud provider – should be “just right.”
In reality, however, the managed private cloud providers haven’t lived up to their potential. In fact, many have seen their growth slow to a crawl and some have failed to deliver the offerings and feature sets they promised. What’s going on? Why are managed cloud providers struggling when they should be the enterprise cloud solution of choice? I believe there are three key factors that are preventing managed cloud services to deliver on their business goals.
Going beyond basic infrastructure services may be the wrong strategy
To justify their cost premium over competing “commodity” cloud providers like Amazon.com and Google, managed cloud providers must deliver more value. To achieve this, some of them are looking beyond infrastructure-as-a-service offerings, bundling in applications they feel have value in the marketplace. That sounds logical. But there are some problems with this approach.
Which services will customers value most? Because every customer is different and has different needs, how can you zero in on a specific application all will be willing to pay more for? For example, one approach might be to include database services in a cloud offering. But which database? One customer might want PostgreSQL while another wants MySQL. Put simply, a “killer app” that justifies the higher price of a private managed cloud just hasn’t emerged.
Perhaps they’re barking up the wrong tree. Instead of adding application services to an infrastructure play, maybe managed cloud providers would be better off delivering a more advanced level of infrastructure services to differentiate their offering from public clouds. That could mean offering a true, fault-tolerant level of availability or a guaranteed service level agreements. Indeed, given some of the well-publicized outages at some of the large public cloud providers, having enterprise-class availability in a managed service portfolio seems crucial to justify their price tag and attract enterprise business.
Lack of strong orchestration capabilities
Achieving economies of scale is essential to the profitability of any cloud service. That means having the ability to orchestrate services, ensuring every bit of data moving around in the cloud ends up exactly where it’s supposed to, at exactly the right time.
Public clouds are typically purpose-built from the ground up, with strong orchestration as a core requirement to enable massive scaling. But that approach is cost-prohibitive for many private managed cloud providers. Instead, they usually assemble their clouds using off-the-shelf solutions from a variety of vendors. Unfortunately, robust orchestration at scale is just not easy to accomplish with the available off-the-shelf solutions; the needs are just too complex. And if you’re going to layer on advanced infrastructure capabilities, as we just discussed, or have highly specialized business or regulatory requirements, the challenge is even more difficult.
To meet that challenge, managed cloud providers need a whole new class of automation and management tools that can work in concert with the available orchestration tools to deliver the functionality they need, at scale. This automation enables them to orchestrate resources dynamically, when and where they need it. Why is that important and how does it add value?
Consider fault tolerance. Maintaining fault tolerance in a traditional on-premise data center usually means deploying the application on hardware that is fully redundant all the time, forever. That gets costly quickly. With dynamic orchestration in the cloud, it’s possible to deploy an application with fault tolerance only when that level of maximum availability is needed. Then, the application can be throttled back to a non-FT infrastructure when fault tolerance isn’t required – completely seamlessly with no interruption of service. A managed cloud provider could offer fault tolerance as a “premium service” that customers pay for only when they need it. That saves customers money, optimizes utilization of service provider resources, and provides the availability required, when it’s most critical. That’s the power of dynamic, automated orchestration.
Too much costly proprietary infrastructure
It is important to realize that many of these struggling managed cloud providers are built upon older, managed hosting business models. This means that they are also struggling to balance newer cloud technology offerings with legacy business structures, tooling, processes, and skill sets that are costly and inflexible. Using legacy data center technologies with high license costs, support fees, and other ongoing expenses may not have been a huge problem for their existing customers. But that is likely to change. To attract and retain customers in the future, managed cloud providers will need to adopt new approaches to solving old problems.
Open source cloud technologies like OpenStack, Linux and Kernel-based virtual machine offer an alternative to traditional approaches for many core functions, eliminating the “proprietary software tax.” But that’s not enough. Cloud service providers need to adopt other new technologies that leverage the flexibility of these open source frameworks, maximizing their efficiency, reliability and automation. This can enable them to deliver their IaaS offering at a lower cost or deliver a premium level of service, such as like fault-tolerant capability like that described above, at the same cost. That’s something that would be too costly to justify with their existing proprietary infrastructures.
It’s true that open source cloud technologies are still taking shape, with some of the most advanced capabilities not yet fully mature. But if managed cloud providers don’t begin to adopt new, more efficient approaches to delivering services, they may find themselves at the wrong end of the competitive stick when their customers discover a less costly alternative.
Embracing change is imperative
Despite the headwinds keeping providers of managed cloud services from fulfilling their potential, the market demand remains strong for private, secure, enterprise cloud offerings on an outsource basis. But, the price pressures imposed by public clouds are forcing managed cloud providers to reevaluate how much of a premium they can charge for their services. To close the gap between what it costs and what the customer is willing to pay, managed cloud providers will need to adopt creative approaches to delivering services.
Cloud providers need to think strategically about what services they need to provide to achieve a premium in the minds of customers. To deliver those services profitably, they need to rethink how they’re delivering services to ensure their approach is as efficient and flexible as possible.
Managed cloud providers who successfully close this “value gap” by embracing new ideas, new approaches and new technologies will have a tremendous advantage as the marketplace for these services continues to mature. Indeed, once their enterprise customers realize they could be getting more and paying less, cloud providers slow to make the shift to new ways of doing things may find themselves out of work.
Editor’s Note: In an attempt to broaden our interaction with our readers we have created this Reader Forum for those with something meaningful to say to the wireless industry. We want to keep this as open as possible, but we maintain some editorial control to keep it free of commercials or attacks. Please send along submissions for this section to our editors at: dmeyer@rcrwireless.com.
