The great cloud migration is well underway. According to the Gartner Predicts 2022 report series, “more than 75% of organizations use multiple public cloud services today, and have plans to expand.”
For some, single-cloud architectures may be sufficient in the short term, but sooner or later, most realize their selected cloud service provider does not meet all their long-term needs.
So when we talk about cloud migration, for the majority of enterprises, we are really talking about multi-cloud.
Migrating to the Cloud? Prepare for Disruption
Moving to the cloud is a disruptive process because of the disparity between on-prem and cloud operating models.
Lack of visibility and control and increased security vulnerabilities are two important differences.
This is true for single cloud architectures, and even more so for multi-cloud. Even if you are currently only in one cloud, the trend is toward multi-cloud, so you should prepare for that now.
Enterprises move to the cloud for simplicity and automation, but what they find is a laundry-list of challenges they weren’t prepared for, including an internal skills gap, increased complexity, reduced visibility, security concerns, and multi-cloud sprawl.
In working with our customers, taking surveys, and polling expert partners and employees, we’ve distilled the top five challenges reported by enterprises when migrating to the cloud. And as I mentioned previously, all of these challenges are multiplied in a multi-cloud environment.
The consequence of inadequately addressing these challenges is that organizations entering cloud environments unprepared are left dangerously exposed to unnecessary security risks.
1. Native security stacks are not comprehensive
CSP native security stacks may do 60 or 70% of what you want, but they are not comprehensive.
Gaps in cloud-native security stacks (like L7 or IDS/IPS) create unwanted exposure and unacceptable risk.
In an attempt to mitigate this risk and provide better visibility, many enterprises opt for the insertion of third-party firewalls, which results in increased cost and complexity.
This is true for single-cloud and is multi-true for multi-cloud.
2. Lack of consistent visibility and control
Without a consistent orchestration engine, cloud and multi-cloud are a nightmare, requiring you to approach “build” and “design” differently in each cloud. So you import legacy solutions. But now, incomplete or different tool sets mean visibility is an issue. This is compounded by how fast applications can grow and scale, which also results in shadow IT and sprawl.
Additionally, CSP network stacks and platforms can be black boxes, and getting root-cause analysis (RCA) or deep insight can be time consuming and difficult.
Some new cloud arrivals look around and say, “Hey, this is great. I’m getting all this stuff as a service.”
What they don’t realize until later is that “as a service” involves lost visibility and lost responsibility, because the “service” does not include ownership. For example, application owners get excited about databases as a service until they realize that they can no longer talk to it privately, and it breaks their Zero Trust architecture. And if it breaks, how do they troubleshoot it?
These holes in your support and security story now puts customer trust at risk. In the event of an outage, how can you guarantee that you can restore service?
3. Cloud exposure is greater than on-prem
Cloud migration brings much-easier access to the internet, significantly increasing your threat exposure. Once in the cloud, easier access to the internet makes network visibility and control more critical than ever. Now that those pesky firewalls and DMZs are out of the way, their compromised hosts are ripe targets for data exfiltration and malicious botnet operations.
4. CSP ‘shared responsibility’ really means ‘your responsibility’
Although CSPs say cybersecurity is a “shared responsibility,” the reality is that you—and only you—are responsible for your own security.
You were promised that the cloud would save you money and make you more agile. But now you notice the critical gaps in the security stack, meaning you need to supply your own legacy firewall technology to shore up your defenses. If you don’t, then you’re creating business exposure and risk.
5. New operating environments create skills gaps
Enterprises are moving to the cloud for the promises of simplicity and automation, but end up facing numerous unexpected challenges. In addition to inadequate security stacks, greater threat exposure and limited visibility and control, they also have to contend with the problem of an internal cloud networking skills gaps.
These challenges are magnified and multiplied in multi-cloud, which increases complexity and makes the skills gap even more painful. “Fixing” the skills gap can be an expensive and elusive proposition.
Here Comes Disruption… Brace for Impact
These are the challenges enterprises face as they migrate their workloads of consequence to the cloud.
You may not experience all of them yourself, but chances are good that you will experience some of them as you move, migrate, and build in the cloud. And if you embrace multi-cloud, expect a multiplier effect on these challenges.
The Center of Gravity Has Shifted
My best advice for dealing with these challenges involves your network operating model, or how you conceive of the cloud, relative to the on-prem environment.
It’s a mistake to think of the cloud as an extension of the data center. This approach tries to cast the cloud in the image of the data center, leveraging legacy tools and devices. But when a traditional networking model collides with the cloud operational model, the resulting impact is painful—and expensive. The reality is that once you migrate to the cloud, the cloud becomes your new “center of gravity.”
To optimize your operation, transform your network and security to a cloud-native model.
Start thinking about your world as cloud-first. Innovation is happening in the cloud, not in the data center. Consider the following:
- If comprehensive security is important to you, start looking for cloud-native solutions that embrace the cloud operational model.
- Look for low friction, high velocity, cloud-native solutions that accomplish the same objectives as your enterprise solutions did. Be careful not to make too many compromises. Whenever possible, try to use best-of-breed tools that give you what you want, in a manner aligned with the cloud operational model.
- Look for tools that provide consistent visibility and control, custom-built to mitigate cloud exposure. Avoid tools and devices like generic firewalls that were not purpose-built for the cloud.
- Plan for an inevitable skills gap and build the expertise you’ll require in the cloud. Cloud, and especially multi-cloud, expertise is scarce, and practitioners can be difficult to find and hire. Consider providing advanced training for your team through a good cloud or multi-cloud certification program, such as the Aviatrix Certified Engineer (ACE) program. You can also get the help you need through Aviatrix’s Advanced Cloud Services.
Multi-vendor and DIY solutions to these challenges are not simple. They are complex. They are expensive. And they are difficult to maintain over the long term.
Wouldn’t it be great if a single, cloud-native solution could close all these gaps, in a repeatable way, that works across all clouds?
At Aviatrix, this is all we think about. Learn about our multi-cloud solution to these challenges, and more.