Multi-cloud is definitely a thing. However, it’s not exactly clear what that “thing” is. According to new survey data from database vendor MariaDB, 71 per cent of survey respondents report running databases on at least two different cloud providers today.
Yet when asked what would keep them from going all in on a cloud database, a vendor’s “lack of a multi-cloud offering” ranked dead last. In other words, everyone is doing multi-cloud, but no one knows why.
Which perhaps supports Gartner analyst Lydia Leong’s contention that, “Most organisations end up multi-cloud, rather than intending to be multi-cloud in a deliberate and structured way.” Pragmatism, not dogmatism, rules.
The database is getting cloudy
If you’re bullish on cloud, MariaDB’s survey data will confirm your bias. According to the survey, roughly 89 per cent of those surveyed expect to see at least 50 per cent of their databases running in the cloud by 2025. Nor is MariaDB’s data an outlier here: Gartner goes one step further, projecting that 75 per cent of all databases will be deployed or migrated to a cloud platform by 2022.
As much as I hope this is true (I work for a cloud vendor, after all), I will admit that it seems overly ambitious. Let’s do some math. First, Gartner pegs global IT spending at $3.9 trillion in 2020. Cloud spending, by contrast, will hit $266 billion. I’m no math genius but I think that means just seven per cent of IT spending is going toward cloud this year (and, of course, a fraction of that will be spent on databases).
It’s possible that tens of billions of dollars in database cloud spend will shift from on-premises to cloud workloads within two years, but IT tends to move slowly, particularly with databases. This isn’t to suggest cloud isn’t a big deal. It is. But it’s going to take time.
There are plenty of reasons for organisations to want to move their database workloads to the cloud. According to the MariaDB survey, here are the primary benefits companies already derive from cloud databases:
- Ease of use (24 per cent of respondents)
- Time savings (23.6 per cent)
- Modernisation (21.3 per cent)
- Cost savings (21.1 per cent)
- [Frees up resources to] Focus on other business aspects (9.3 per cent)
Pretty compelling, right? Spurred by these benefits, respondents were asked what would keep them from going all in on a cloud database:
- Security (73.3 per cent)
- Price (46.2 per cent)
- Compatibility (44.9 per cent)
- Scalability (35.2 per cent)
- Migration (33.1 per cent)
- Lack of multi-cloud offering (21.1 per cent)
- Other (1.3 per cent)
It’s interesting that “multi-cloud” ranks last here, despite (as noted above) the fact that 71 per cent report running databases in at least two clouds. Multi-cloud doesn’t seem to be a motivation to buy. Rather, it’s just a fact of enterprise existence.
Enterprise is as enterprise does
Many years ago, Billy Marshall coined the phrase, “The CIO is the last to know.” He was referring to how open source software makes its way into enterprises through developer laptops, but the same phenomenon plays out today with cloud services, even at smaller companies.
At one start-up we first guessed how many SaaS applications we had running; I don’t remember the number but it was small, like 10. When we actually tallied up the true number it was closer to 80.
As noted above, “ease of use” is the primary benefit companies derive from the cloud. This convenience factor opens the door to more cloud finding its way into the enterprise, not less. Leong calls out this longstanding enterprise principle with her typical style:
Multi-cloud is inevitable in almost all organisations. Cloud IaaS+PaaS spans such a wide swathe of IT functions that it’s impractical and unrealistic to assume that the organisation will be single-vendor over the long-term. Just like the enterprise tends to have at least three of everything (if not 10 of everything), the enterprise is similarly not going to resist the temptation of being multi-cloud, even if it’s complex and challenging to manage, and significantly increases management costs. It is a rare organisation that both has diverse business needs and can exercise the discipline to use a single provider.
To Leong’s point, multi-cloud may be a fact of life but it doesn’t tend to be one that companies choose, for the reasons she cites.
But what is an intentional, critically important factor in cloud (and cloud databases) today is hybrid. Asked how important hybrid cloud was to corporate objectives, respondents were very clear:
- Not at all important (2.5 per cent)
- Somewhat important (19.4 per cent)
- Very important (39.2 per cent)
- Extremely important (38.9 per cent)
That is, 97.5 per cent of respondents believe hybrid cloud is at least somewhat important, with a whopping 78.1 per cent saying it’s very important or extremely important.
The reason for this should be clear from the IT spending data I laid out earlier. However much enterprises may want to get to the cloud as quickly as possible, they still have workloads that will stay on-premises for a time, and sometimes forever.
As such, enterprises rightly are much more concerned with getting their hybrid cloud strategy right than worrying about their multi-cloud non-strategy. It may not win enterprise IT any buzzword bingo contests, but hybrid is where the action is.