Much of the focus on cloud computing over the past several years has been simply making the transition. But as IT organizations move beyond running a handful of isolated application workloads, they are starting to discover getting to the cloud is only the first step in what is quickly becoming a career-long adventure.
A survey of 100 IT decision-makers in companies with 500 or more employees conducted by NetEnrich, a provider of an IT automation platform, finds 85 percent of respondents reported either moderate or extensive production use of cloud infrastructure, with 80 percent claiming their companies have moved at least a quarter of all their applications and workloads to the public cloud.
But as more applications start to be deployed at scale in the cloud, a whole range of new concerns arise. Not surprisingly, the survey finds the top cloud computing concern cited to be security (68 percent). But close behind, the survey also finds that other major cloud concerns now include IT spend and cost overruns (59 percent), day-to-day maintenance (36 percent), and root cause analysis and post mortems (22 percent).
Obviously, many of these same issues raise their ugly head in on-premises IT environments as well. But nearly half the respondents (48 percent) say their IT organization is finding the cost of recruiting IT professionals that have right skills to take on these challenges to be an ongoing issue.
In fact, it’s that lack of available cloud skill set that results in so much spending on shadow IT projects being funded by line of business units, says NetEnrich CEO Raju Chekuri. According to 56 percent of the survey respondents, 20 percent to 40 percent of enterprise technology funding is being spent outside IT’s purview.
“The relevance of the IT team comes down to skill sets,” says Chekuri.
Unfortunately, adoption of best DevOps practices for managing cloud applications is uneven at best. Only 23 percent of respondents said their organizations have completely switched to DevOps, and 18 percent have not made the shift at all, even though 86 percent of respondents claim to have rearchitected some, or all, of their applications to use cloud-native services. On the plus side, a full 59 percent of respondents say they use some tools for continuous integration and continuous code deployment.
A separate survey of 336 developers, web product managers and DevOps practitioners published by SolarWinds, a provider of IT management tools, finds most organizations are simply too swamped by basic troubleshooting of applications to make any meaningful transition to DevOps.
As organizations transition to the cloud, the most pressing issue may have little to do with the underlying technology, says Siki Giunta, managing director of cloud and infrastructure at Accenture. Organizations that tend to succeed most have aggressively shifted toward creating cross-functional teams that are organized around the delivery of a specific application or service, says Giunta.
“It’s really a mindset change,” says Giunta.
As part of that mindset shift, many organizations are opting to rely more on third-party managed service providers such as Accenture to manage their cloud operations, which in turn frees up dollars for building applications, says Giunta.
Integration Challenges
Cloud computing challenges, however, are not simply limited to cybersecurity, cost and process. A separate survey of 100 IT decision makers conducted by Cleo, a provider of integration software, reveals cloud integration challenges result in losses that can be as much as half a million for enterprises annually. Over half the respondents (57 percent) admit that poor integrations and lack of resources contribute to hundreds of lost orders per year, leading to annual revenue losses in the range of $250,000-500,000.
Primary cloud integration challenges surfaced by the report include the fact that onboarding new business partners and customers is too complex (63 percent) and that they lack the skilled resources needed to build and manage integrations between systems, applications and partner ecosystems (29 percent).
The biggest impediments to achieving that goal are all the legacy systems, most of which still run on-premises, that organizations still depend on, says Tushar Patel, chief marketing officer for Cleo.
“It takes too long,” says Patel.
The Cleo survey finds 38 percent of survey respondents lack confidence in their capacity to scale in support of integration initiatives, while 22 percent say legacy technologies cause significant delays in generating new revenue. A full 81 percent said replacing legacy systems will better support emerging business initiatives.
The biggest problem, of course, is that too many organizations still don’t have a cohesive strategy when it comes to the cloud, says Peter Kraatz, director of cloud and data center transformation for Datalink, a division of Insight. Too often, organizations are haphazardly moving applications to the cloud.
A recent survey of 200 IT leaders conducted by International Data Corp. (IDC) on behalf of Datalink finds just 28 percent of the survey respondents say they have documented and communicated a cloud strategy. Determining which workloads should move to the cloud is most often ranked as the top challenge in executing a cloud strategy, running neck and neck with new tools needed, change management, and choosing cloud deployment models.
“A lot of organizations simply underestimate the size of their IT estate,” says Kraatz.
Over half (51 percent) of survey respondents admit the challenges they’ve experienced have prompted them to stall or abandon some IT transformation initiatives.
The survey also finds most organizations say they have a “cloud first” policy, but there is confusion over the definition of “cloud first.” Those organizations that define “cloud first” as hybrid cloud strategy have made the most progress with IT transformation versus “cloud only,” notes Kraatz.
But many organizations also fail to appreciate the challenges associated with managing data across a hybrid cloud computing environment, adds David Richards, CEO of WANdisco, a provider of tools for automating the movement of data between IT environments.
“For some customers, you’re talking about petabytes of data,” says Richards.
Dealing with that much distributed data requires adopting a global file system to automate the management process, say Richards.
Given all the complexities involved, it is little wonder that so many organizations have found moving to the cloud a much bigger challenge than they might have originally anticipated. That doesn’t mean those organizations will be retreating from the cloud any time soon. But it does mean that there is now much greater appreciation for all the technology and process challenges involved in making that transition.
