Despite some high-profile hiccups along the way, demand for cloud services continues, and it appears as though there’s no place for it to go than up. Analysts forecast ongoing growth for such services, telcos and others are snapping up cloud businesses at a rapid pace, and rumor has it that even Apple is funneling some of its energies into launching a cloud-based offering.
Sixty percent of 3,000 global CIOs that IBM (News - Alert) recently surveyed said their organizations are ready to embrace cloud computing over the next five years as a means of growing their businesses and achieving competitive advantage. (That’s nearly twice the number of CIOs who pledged their allegiance to cloud computing two years ago.) The IBM research also indicates that seven out of 10 CIOs in the U.S., Japan and South Korea, and 68 percent in China, now identify cloud as a top priority. Meanwhile, Ovum (News - Alert) says 63 percent of multinational corporations in the Asia Pacific are using at least one type of cloud service, which the firm says defines as including networking, communications, applications, corporate IT systems, data management, security and backup.
As for other businesses, Microsoft reveals that about 39 percent of small and medium-sized organizations expect to adopt cloud services – such as collaboration, data storage and backup or business class e-mail – within the next three years, compared to the 29 percent that are paying for such services today.
That helps explain why Gartner forecasts that by the end of 2011 cloud-based services will account for almost a quarter of the overall hosting market, excluding co-location and mass-market hosting.
"Cloud computing and hosted managed services will without doubt become the primary method of IT delivery to organi[z]ations large and small,” says Keith Bates, chairman of the Cloud Computing Centre.
That said, it’s no surprise that many communications businesses are entering, or expanding their efforts, relative to cloud services.
Indeed. Fresh off from closing its deal with Qwest (News - Alert), CenturyLink Inc. this spring announced plans to buy Savvis Inc., a leader in the cloud infrastructure as a service and web hosting space, for $2.5 billion. That came in the wake of Verizon completing its $1.4 billion purchase of Terremark Worldwide Inc., as well as Cincinatti Bell’s CyrusOne (News - Alert) acquisition, Time Warner Cable’s NaviSite buy, and many others.
As the 2010 Gartner Magic Quadrant study on this sector notes: “This is a time of both great opportunities and great risks for the service providers in this market. New entrants are altering the landscape, and established hosters that previously lagged the market have been able to make bold investments in an attempt to catch, or even overtake, more established competitors. Most providers are investing aggressively in technology innovation and exploitation, and we believe that mergers and acquisitions will become more commonplace as vendors seek to decrease their time to market, obtain engineering expertise with new technologies, and build market share.”
Of course, it’s not just web hosting businesses and telephone companies that are moving on the cloud opportunity. Amazon, which got its start as an online bookseller, is considered the leader in cloud services, with more than 40 percent market share in 2010. And reports indicate that another A-list company – Apple – may not be far behind.
Apple didn’t respond to INTERNET TELEPHONY’s requests for an interview for this piece, but
GigaOm reports that Apple may have paid four and a half million dollars to purchase the domain name iCloud.com from Xcerion, which had used the name to host its cloud-based storage system. Xcerion recently changed its domain from iCloud to CloudMe, according to the same piece. That and the fact that Apple recently hired former Microsoft data center guru Kevin Timmons are clear indicators that the inventor of the iPhone is getting serious about the cloud, some believe.
This all bodes well for the future of cloud services. But just as such offerings are gaining broad acceptance, incidents like Amazon’s recent web services outage have given greater credence to concerns about the reliability and security of cloud communications.
In mid April Amazon experienced a multi-day outage resulting from a configuration problem during a regular capacity upgrade effort in a data center in northern Virginia. The company said that disconnected both the primary and secondary network simultaneously, leaving the affected nodes completely isolated from one another. And that took down a bunch of websites, including the sites of popular social networking sites Foursquare, HootSuite, Reddit and Quora.
Naturally, Amazon apologized and promised to “do everything we can to learn from this event and use it to drive improvement across our services." And, naturally, everybody else examined and opined about what the event means for the cloud service space and what it should mean for customers using or considering such options.
What is clear is that applications will live in the data center – whether those applications and that data center are owned and operated by Amazon or another cloud service provider, or by the company itself – and with that comes a certain amount of risk. Whether that risk is any higher when the application or infrastructure is outsourced to a cloud provider is what is at issue.
"I think it's a mistake to expect cloud computing to be foolproof,” Mike Eaton, CEO and founder of Thousand Oaks, Calif.-based Cloudworks, tells INTERNET TELEPHONY. “It's not. It's still IT, and IT is complex, and complexity is the enemy of reliability.
“The question is whether cloud computing is more reliable than other forms of IT delivery,” Eaton continues. “And it is. When faced with the option of cloud computing or bringing IT in house, I think there is no question that cloud computing, dollar for dollar, is significantly more reliable than an in-house alternative."
Thomas Howe of consulting firm Embrase, which produces the Cloud Communications Expo, seems to agree. As quoted in a recent TMCnet article, Howe says “The question isn’t: Is the cloud safe? The question is: Is the cloud safer than what I can do? For nearly all companies, Amazon wins that battle.”
Howe went on to suggest that companies concerned about reliability might limit the damage from such outages by diversifying their cloud services provider supply chain.
John Engates, CTO Rackspace Hosting, adds, “Companies should do their diligence with regard to the promises made by their cloud provider and the SLAs in place to back them up. Transparency is key. If the provider is not transparent with their architecture, it's impossible to fully understand if what you're building is going to be resilient and highly available on top of that provider.”
And Jake Sorofman, CMO for rPath, comments: “The recent Amazon EC2 outage was a highly complex event, but it was ultimately triggered by a single manual configuration mistake in a multi-step operational procedure. Clearly, the oft-quoted statistic that 80 percent of outages are due to config mistakes holds true, even in the largest and best-run shops. The answer, as always, is automation – multi-step procedures must be sequenced in an orchestration system, while low-level config changes must be model-driven and pre-validated.”
While it was a configuration change that led to the Amazon outage, what happens within the data center is just a part of the reliability equation, as several sources note.
As Alan Murphy discusses in the Virtualization Reality column in last month’s issue, “Nowhere does the WAN play a stronger role than in off-premises cloud computing.”
The F5 Networks technical marketing manager and regular INTERNET TELEPHONY magazine columnist went on to write: “By design, off-premises clouds are at the mercy of the WAN because they reside outside the walls of your data center and off the LAN. But also by design, and one of the major differentiators between cloud deployments and traditional hosted environments, is in most cloud architectures the off-premises infrastructure needs to be tied to your on-premises infrastructure; there needs to be some connectivity over the WAN between your data center and the cloud provider. Not only do you have to deal with optimizing and managing data to and from your cloud provider (and users), but now you have to bring those issues in-house.”
That would seem to be a tricky one for Amazon given, as Forrester notes, the Elastic Cloud Compute platform operator lacks the co-location, dedicated connectivity and servers, managed services and SLAs that many of the leaders in the cloud services/web hosting space have in their portfolios.
Indeed, companies like Verizon Business consider their ownership of wide area network assets as a differentiator for them in delivering and supporting cloud-based solutions.
Peter Konings, director product marketing at Verizon Business, points out that as customers put more of their applications in a data center in the cloud, the underlying network is a key component in the solution. That’s one of the reasons, he says, Verizon continues to invest heavily in providing redundancy on its MPLS network. The company employs mesh network technology to allow restoration of its network even in catastrophic situations like in the tsunami in Japan, he says.
“MPLS is not the only game in town, but it has the performance of any game in town,” Sweet says, referring to the traffic prioritization via QoS features of MPLS.
Konings adds that Verizon Business customers are requesting investment in more diverse nodes within existing regions or existing cities. Customers that previously were comfortable using a diverse node in a connection into a city a few thousand kilometers away probably no longer find that acceptable, he says. Now customers want redundancy within a city region. Even when outages occur, he continues, customers are still looking for very reliable latency.
Then there’s the public vs. private connectivity issue. As part of the Verizon Business computing as a service solution, which relies on multiple Verizon data centers in Asia Pac, Europe and the U.S., the company provides customers access to servers, storage, etc. When the service launched Verizon gave customers the choice of using either public IP connectivity or private connectivity. Konings says that 75 to 80 percent of its customers are going for private connectivity because it offers end-to-end solution control and enables Verizon to offer stricter service level agreements.
When customers are moving to public cloud-based applications like Gmail and Google Docs, he continues, they need to consider the impact of that change to their wide area network services. In such situations, he says, customers tend to put a lot more data on the public IP network, so they might want to look at using private links for the most important applications to control quality.
“These are things which I think we will continue to see more and more,” says Konings.
Edited by Stefania Viscusi