Bracing for a future that involves AI and ever-increasing data sets, CIOs face great cultural challenges.
Multi-cloud strategy: 5 challenges you’ll face
Using multiple cloud providers? Let's examine advice on the problems that arise – and how to beat them
Many organizations are moving full steam ahead with multi-cloud strategies. That doesn’t mean there won’t be some speed bumps along the way.
A successful multi-cloud strategy – multi-cloud meaning you use more than one cloud service, from more than one cloud provider – must account for certain challenges.
[ Are you ready to discuss cloud security concerns? Get our concise guide and learn from the experts: Hybrid cloud security: 5 questions skeptics will ask. ]
As multi-cloud strategies continue to mature, we asked a range of experts to weigh in on the problems multi-cloud IT shops commonly encounter – and must solve to maximize the potential benefits.
These issues all can be solved: They should not be viewed as deal-breakers for your long-term multi-cloud strategy, but rather as opportunities for learning and proactive solutions. Let’s dive into those challenges, in no particular order.
1. Data governance and compliance
Data jurisdiction drives some organizations’ move to a multi-cloud strategy: With multiple clouds and data centers in multiple geographical regions, you can achieve both flexibility and compliance.
That doesn’t happen automatically.
Dave Dozer, business systems consultant at Algorithm, Inc., says that one of the bigger challenges is understanding where, exactly, your data physically resides. (Dozer notes that this problem may be especially tough for smaller and midsize companies.)
“Moving applications to the cloud doesn't eliminate the need to meet certain data governance requirements,” Dozer says. “And in a multi-cloud environment it may be easier to [make a mistake] and not realize that a specific application or piece of an application is running in an unapproved environment.”
Dozer points to ITAR compliance as one example. (Of course, there are plenty of others. GDPR, anyone?) If you’re required to meet those regulations, that would mean your data can’t be accessible to non-U.S. citizens or stored outside of the United States.
“These sort of compliance issues can cause headaches for IT leaders when deploying in a multi-cloud environment because each instance must be scrutinized,” Dozer says.
Again, this isn’t a dealbreaker – but it’s an issue IT leaders must prepare to solve with the proper tools for visibility and monitoring depending on the extent of their regulatory burden.
This may also be a reason why a multi-cloud strategy is increasingly likely to go hand-in-hand with hybrid cloud: By integrating your various public and private cloud environments with your on-premises infrastructure, you may be better positioned to address some of the data residency complications that can arise.
[ What’s the difference between hybrid cloud and multi-cloud? ]
“In today’s data-driven culture, it is often a requirement for multiple different solutions to have access to the data of several other solutions,” says Chris Hansen, cloud practice lead at SPR. “The performance and complexity of creating secure access to data cross-platform is less controllable in a multi-cloud strategy then it is in a hybrid cloud or single provider strategy.”
The single provider approach may be going the way of the dodo. Research firm IDC has predicted that 90 percent of enterprises will be using multiple cloud services and platforms by 2020.
Hybrid cloud, therefore, is likely to be an increasingly appealing architectural choice for proactively managing data in a multi-cloud environment.
2. Multiple skill sets and vendors to manage
Rich Murr, CIO at Epicor, notes that multiplying your cloud environments typically means multiplying what’s required to effectively run those environments over the long haul. That’s true both for your internal teams and your external resources.
“The higher the number of IaaS, PaaS, and SaaS solutions leveraged, the higher the number of [in-house] technology skills that must be developed and maintained, and the greater the vendor management burden,” Murr says.
Murr’s advice? Strive for common denominators as much as possible.
“While multi-cloud is practically inevitable for larger IT teams, especially those responsible for developing and operating both corporate IT solutions and customer products, IT leaders should strive for commonality whenever possible,” Murr advises.
His examples of commonality across diverse and distributed environments include: Operating systems and system administration tools, development languages and frameworks, integration technologies, and cloud management platforms.
3. Software development and delivery
One particularly important – and sizable – place you can look for such common ground is your software pipeline.
“One obvious challenge with going multi-cloud is the complexity it may introduce to application delivery,” says Amir Jerbi, CTO at Aqua Security. “Organizations should avoid cases that require reconfiguration or cloud-specific adaptation of applications, as well as any feature disparity. If moving an application between clouds requires significant work every time, or results in the application behaving differently when running on different clouds, then you’re losing many of the efficiency gains.”
[ Containers can play a key role in solving this problem. Read our post, 5 advantages of containers for writing applications. ]
This also reflects the importance of standardization – standardizing on tools and processes is a common characteristic of effective DevOps teams, for example – to the scalability and flexibility of multi-cloud environments.