Cloud and Multi-Cloud Revisited
The cloud has revolutionized how business is conducted, and I believe this is something we all can agree upon. However, by using the cloud, has it made things easier for us? Yes, indeed in some ways, but in some ways perhaps not. No doubt it has made our environments more flexible, dynamic and agile, which is all a good thing. But it also has made our operational budgets (Opex) go up, while our capital budgets for IT (Capex) go down. It means we might spend more overall (cloud is not always cheaper) to provide the same IT services as when we had technology on-prem. Is that OK? Yes and no, depends up your perspective. What the cloud has done is open up a different set of complications than we had before, which is even further exacerbated by muli-cloud environments. That’s what I want to talk about—some would say lament about—this month.
So what is multi-cloud? At its core multi-cloud is using different cloud services to perform various functions, all under the same architectural umbrella. Sounds pretty simple and straightforward, right? However, when you have different cloud services that must “work” and be managed under the same umbrella—but with various providers enabling services not hosted at your location—what could go possibly go wrong? There’s complexity brought to the table in using multiple cloud providers, not dissimilar from when you use different providers/manufactures for on-prem technology.
Many will use one provider for IaaS, such as AWS or Azure by Microsoft, Microsoft for PaaS, others for SaaS, O365, Salesforce, etc, and others for backup such as Datto (BU/DRaaS). That’s just a few. We all know there are many different cloud providers out there that offer these (seemingly the same) services. Navigating the landscape of cloud solutions can be daunting. I think that’s why you see firms—and people— getting cloud fatigue. Just too many choices out there to consider. That’s also why you see firms choosing to go with larger providers, such as Microsoft and AWS, because they are well known, many of their colleagues use them, and seem easy choices. That’s not a bad reason to choose them; but it shouldn’t be the only reason.
Many of the cloud providers offer best of breed solutions for specific industries and specific needs, which is a very good thing. It includes different price ranges and features available, and many focus on specific niches or even solutions for particular industries. Again, choice is good; too many choices may not be. Let the market dictate what it wants/needs and the good/stellar cloud providers will remain and flourish. Many will not; we see it all the time.
With all of these different cloud solutions that firms have access to and use (which is precisely what multi-cloud is), what are the security implications? Well, you can imagine what this means for security and governance; it’s the sort of thing that keeps CISO’s on their toes. Multiple cloud providers give you services that are not at their location by yet have access to your data and your environment, at the same time. Again, what could go wrong? Well, quite a bit, actually, as security becomes more complicated and the potential for problems is magnified. When something happens with your data in your cloud environment, who’s to blame? Who’s responsible? What does that means for governance and board/shareholders in the their confidence in a firms ability to successfully use the cloud?
Multi-cloud also opens a host of questions—some similar to on-prem environments, and some unique to cloud. An example is application rationalization services. Firms always have had to make sure that their applications will work if whatever new environment is used, and must be ported to do so. Windows to Linux, x86 to Power, etc. But what happens when you want to move from on-prem to the cloud? It’s not just a simple lift-and-shift as many would think. In fact, lift-and-shift by itself is not a good value prop for the cloud. It doesn’t provide the best performance (it could take a hit) unless the applications are re-tuned to work in the cloud. Sounds simple, but it’s not, as we all know.
Then there’s the overall cloud strategy. What design and architecture is to be used? What is the migration strategy for each platform and application? How do you manage and provide automation? Then there is backup of your environment and host of other considerations. What happens if your cloud provider goes down? How do you get up and running in another cloud environment or an on-prem environment? What is the best cloud provider for that specific workload? How do you manage and control costs, which is a big wake up call for a lot of firms, where they are spending a lot more money than they thought on cloud. What about integrating your cloud workloads with other cloud workloads and other services, including your on-prem services. How do you integrate them all? And what about bandwidth? The cloud requires robust and predictable data and networks, as we all know. Take all of these together and there is a lot to work through (and be concerned), and this is just some of the considerations. But that’s cloud for you.
So what does it all mean? Whether your workloads or on-prem or in the cloud, there is a lot to consider. The cloud brings much and many things to consider that in some cases are the same are your on-prem environment, but also many other areas as well. You’ve heard me say it before, but this is where a good partner can help. They can help you make sense of all of this, which is an important step.
Done right, your cloud services can give you what you need to be successful under one unified strategy and architecture; done wrong it can be jumbled mess. None of us wants a jumbled mess with our cloud environments, or anything we do in IT, for that matter.
Until next month.