While we monitor the market as a whole, cloud adoption statistics are indicators of market share among the large providers. The ParkMyCloud platform sees a very large volume of data flow through it each day, month, and quarter, and therefore affords an interesting and helpful perspective on our users. When this in turn allows you to examine their usage preferences of other third-party services it can be downright enlightening.
For this post, rather than examining the granular detail of specific user preferences for certain products and services, I thought it might be interesting to roll the numbers up and see what this says about the world of public cloud as a whole. In particular, given we are now at the end of earnings season (see our recent post here ) we thought it would be interesting to compare what we see in our customer base. While obviously we only have a tiny percentage of the overall user base of public cloud, in recent years it has increasingly been my belief that it is fairly representative of the overall market. In fact I would go as far as to say that what we have observed in our data often appears in the public announcements some months later. The big trends such as increased usage of very short lived instances (especially for data analytics workloads) or increased use of custom instances have caught our eye only to be affirmed more broadly by the market.
Some of the things we look at each quarter include:
Relative changes in the proportions of customers exclusively using one of our supported cloud providers ;
Relative changes in the proportion of customers using multiple clouds (typically two or three different providers); and
The total number of accounts each customer has with each provider.
Over the last year or so it has been interesting to see the shifts occurring in the first of these measures – customers making exclusive use of a single cloud. Putting to one side an obvious caveat which is that customers could have other cloud accounts not brought into the PMC platform, we have observed that some 95% of users are exclusively using a single cloud. There have been some shifts in the relative proportions using either AWS and Azure, AWS and GCP and Azure and GCP but the numbers here are so small compared to those using a provider exclusively, that it is hard to draw any strong conclusions.
Figure 1: Changes in Customers Making Exclusive Use of Cloud Provider (Source: PMC).
However, once again putting the overall representativeness of the PMC user base to one side, we can without doubt see some meaningful changes over the last six quarters in which clouds are being exclusively used by our customers. The chart shown in Figure 1 above, shows the relative changes over the last 18 months, with Q6 being March-May 2020. It is therefore likely that we picked up some of the COVID-19 related shifts, but will see more in the coming quarters.
To show these changes I have rebased the data (Q0) and then looked at the relative changes over the period. So for example, you can see that exclusive Azure users grew their footprint amongst our customer base by some 9.2% by Q5 and ended the period up 6.2%. There is a clear upward trendline for Azure during these last six quarters, versus AWS and GCP which are showing a flat to a slight downward trajectory. As mentioned above the proportion using multiple clouds has stayed fairly static.
It will be interesting to see how these cloud adoption statistic trends play out. Based upon what we are hearing anecdotally from our customers, there has been a lot of growth in the market for Virtual desktop infrastructure (VDI) in the move to remote working, and that Azure have been the largest beneficiary of the shift. With employers increasingly alerting staff to the possible realities of home working throughout the winter months, we think it likely that the trend continues.
What we do know from the earnings numbers is that the growth in cloud revenue numbers is slowing down for all three providers with the steepest declines being reported by AWS (although actual earned revenue is still increasing for all three). In such an environment, competition for market share is likely to get even more intense and so monitoring these shifts is likely to become even more important to track.
Over the past five years, we’ve seen the challenges of cloud computing evolve – but ultimately, their core needs are the same as ever.
It’s interesting to experience and observe how these needs get translated into products, both our own and others. Depending on company growth and culture, Build ↔ Measure ↔ Learn cycles can continue to turn in a rapid fashion as ideas get adopted and refined over time. Or, they can become bogged down in supporting a large and often demanding installed base of customers.
In a few short years, tools built for optimizing public cloud have evolved into a number of sub-segments, each of which in turn has developed to meet customer needs. In part, this reflects a predictable maturation of enterprises using cloud computing as they have migrated from on prem to public or hybrid cloud, and adopted best practices to enhance performance and security while tackling overall growth in monthly spend.
How This Year’s State of the Cloud Report Stacks Up
Flicking through various analyst reports and “Cool Vendor” white papers, it’s fascinating to see how quickly cool becomes uncool as industry needs develop. Being a social scientist by training, longitudinal panel type surveys always grab my attention. RightScale/Flexera’s annual customer survey ticks a few of these boxes. No doubt the participants have changed, but it likely provides a valuable source of data on customer needs and challenges in cloud computing.
You do not need to go back to the first RightScale survey in 2013 to see some big changes. Even when comparing the 2016 to the 2020 survey in terms of company priorities in the cloud, it’s hard to believe that just a few years ago, the number one challenge regardless of maturity was related to a skills/resources gap. These priorities were followed by security and compliance issues, with the focus on cost optimization being a lower priority and then only for those at a more mature state. Roll forward to 2020 and cost management is now the number one cloud initiative in all but the most recent adopters of cloud where it sits at number two. Interestingly, security seems to have dropped off the top-five list. Governance has held on, although likely headed the same way. Conversely, cost optimization now sits atop all other initiatives.
Why Is Cost Optimization Still #1?
What seems to be apparent when reading between the lines of such reports and when talking with customers is that unlike migration, security, and governance, there are still some large holes in companies practices when it comes to optimization and reducing cloud waste. Despite a plethora of tools on offer in 2020 that offer to bring visibility and cost management that the overall cloud waste number is actually still growing as infrastructure grows.
More money has been spent tackling security and governance issues – and these challenges in cloud computing need to be dealt with. But cost optimization can deliver ROI to free up budget to deal with these issues.
In the wake of COVID-19, finance teams across the world will now be sharpening their pencils and looking more aggressively at such measures. While cloud spending may rise, Gartner and IDC have both forecasted overall IT spending to drop 5-8%
Yes, You Can Optimize Now
As with security and governance, a mix of human behavioral and business process changes will be required, both of which can be supported by effective tooling, both native cloud provider and 3rd party ISV tools. Incentives to implement such changes are likely to be higher than in the past, albeit in a more cash-constrained world where low cost, ease of use, and most of all, quantifiable ROI will be prioritized. It has always appeared to me somewhat oxymoronic when I hear promises of reducing cloud waste through the use of expensive cloud management tools that charge based on a percent of your spend.
I foresee a wave of low cost, multi-cloud, and simple to use tools emerging. These tools will need to demonstrate a rapid ROI and be built to be used across engineering and operations (not just in the offices of the CIO/CTO/CFO) to ensure the self-service nature of cloud is not disrupted. A similar pattern will emerge as these tools become part of day-to-day cloud operations where cost optimization is part of the cloud culture. With this the need for specific cost optimization initiatives should be replaced by a new wave of needs, like application resource management.
As we lurch into the roaring twenties we thought it might be interesting to consider what lies ahead in the ‘exciting’ world of cloud governance. We are unsure exactly when ‘shadow IT’ reached ‘peak sprawl’ but it was likely at some point in the middle of the twenty-tens. What seems certain is that when it comes to IT governance there is still the same need to balance the benefits of agility and speed which come from decentralization, against key business risks be they security and/or cost management.
One thing for sure is that what is encompassed within cloud governance very much depends on where you sit within an organization. Microsoft has produced some interesting content on these different perspectives, which they have boiled down into a notional ‘Five Disciplines’. From our own perspective, we probably think mostly around cost management and cost optimization. Obviously, if you sit within a security related function within a company or are a vendor of security tools, cloud governance means something quite different. The other factor impacting perspective is where you stand on the so-called ‘cloud journey’. If you are still working on migrating your first workloads to the cloud you will have a completely different outlook than if you have been in the cloud for the last 10 years and built your entire business model from the ground up in the public cloud.
Now that we are in 2020, what does this all mean? The cloud world is full of predictions but one often cited that caught our eye is that in 2020 some 83% of enterprise workloads will be in the cloud with approximately half of these being in the public cloud (AWS, Azure and GCP for example). The growth in public cloud over the last decade has been enormous and with it a management task that has moved beyond a scale that humans can manage. Automation has been part of the cloud since its inception, but the move to automated governance has begun and without a doubt will continue to accelerate in the coming years. Be it automated, from cloud guardrails which prevent misconfigurations which enable malicious attackers to penetrate what was considered well protected systems, to automated cloud cost control which automatically schedules resources to be available when required (and off when not) or adjusted to the rightsize to meet the needs of the workload. It’s also not just the infrastructure layer that’s going to get automated as new tools emerge including application resource management, which enables the entire application stack to be automated using software.
In reality, most of what is termed ‘automation’ in the world of cloud governance in 2020 are in truth recommendations which are then manually implemented or for those who are able to align with internal process some kind of semi-automation such as the use of policies. These often still require sophisticated workflows, approval processes and sign-offs from operations and business owners. Few organizations have moved to fully automated governance actions were, in essence, the machines are being used to manage machines. Just as with the move towards autonomous vehicles, driver augmentation via adaptive cruise control, lane-centering, etc. is now considered almost standard on new cars, and so is at least some level of automation in governance is becoming a standard requirement. Being delivered a list of hundreds of recommendations in the last decade was considered a vast improvement on the status quo. In the next decade, these recommendations will likely increasingly become invisible as infrastructure optimization is managed in an ongoing and continuous manner and will require little or no human input.
The range of governance tasks to be automated is also likely to grow. We can already observe the way cost management is increasingly being automated and our own customers are getting comfortable with more ‘set it and forget it’ automation processes based on policies they define. Teams anxious about cloud security are turning to a growing market of automation tools that cover off monitoring, compliance, and threat management and remediate these issues in real-time.
For sure there is a lot of headroom when it comes to automating governance and we look forward to seeing where we land by 2030.
As the end of the year approaches, and we look ahead at what the 2020 tech trends promise to have in store for the cloud, we can’t help but also reflect on what the past years’ trends have foretold and given us thus far. As we enter 2020 we are not only entering a new year, but also a new decade, so it’s doubly interesting (and fun) to sit back and ponder on what the year and decade ahead might hold.
Before summoning the oracle and thinking ahead specifically on the future of cloud management, it’s worthwhile looking back at the big picture over the last decade to give us some sense of what we have to look forward to. Let’s start with the proverbial ‘gorilla’. AWS was founded in 2006 and had reached annual revenues by 2010 of ~$500MM. Not bad growth from a standing start and a growth trend which continued throughout the decade. With Wall St. estimates of some $50B in revenue for 2020, this means 100x growth. That is quite simply incredible and with growth last year at 35% year-on-year, this AWS growth doesn’t look like it will stop..
2010 Cloud Prediction
Amazon Cloud Revenue Could Exceed $500 Million In 2010, CRN (2010)
Growth among Microsoft Azure and Google Cloud Platform has also not been too shabby but AWS has held (and in many ways strengthened) its dominant position over the last decade.
Due to the wonderful archival powers of the internet, finding white papers on the future of cloud from a decade ago is no more than a few clicks away. Rather than try and summarize them here, it’s worth reviewing yourself – one that’s worth taking a look at is Microsoft’s The Economics of the Cloud (2010). Some parts were right, some wrong, but the key point was that: “cloud services will enable IT groups to focus more on innovation while leaving non-differentiating activities to reliable and cost-effective providers”.
On this particular point, it’s hard to argue that as a result new companies and new business models have been realized in ways previously not imaginable. Be it in the world of the sharing economy, Uber, Lyft, Airbnb, etc or the myriad of other cloud powered unicorns which have been built over the last decade, cloud infrastructure has been a huge enabler of growth.
As interesting as it would be to speculate on where the cloud industry might be headed to by 2030, (AI-Cloud, IOT, Blockchain, Space Cloud Computing, etc) and so that we keep our feet firmly on the ground, the trends we at ParkMyCloud feel qualified to comment on are somewhat more modest and closer to home.
Cloud Management – What we are seeing here is demand from customers for a more consolidated view across their multiple cloud accounts. In 2019 we saw a lot of our customers going mainstream in the multi-cloud world and trying to integrate a mix of cloud-native and third-party tools to provide actionable insights and more importantly actions. The companies building Cloud Management technologies have grown over the last decade but in many ways, it remains a small and no unicorns have yet emerged. We think this will change in the coming decade as the management of cloud infrastructure in all aspects (technical, economic, etc) has reached such a scale it requires non-human intervention and coordination.
Multi-cloud – Multi-cloud truly arrived in 2019 and we believe it will grow in 2020. Most organizations now use multiple clouds and among our customer base, there appears to be less concern about vendor lock-in. We also increasingly see specific clouds being used for specific purposes, so, for instance, data analytics workloads utilizing one particular cloud provider, whereas development and production sit on an entirely different cloud.
Automation – Building on what we see happening in the world of Cloud Management, the demand for automation across the technical and economic management stack is growing. Companies are getting more comfortable with semi-autonomous modes and in some cases moving to full-blown automation. Many have drawn the comparison with the 1 to 5 scale used in the field of autonomous vehicles and we like this analogy. With many now operating at level 2(Partial) or level 3(Conditional) we see this continuing to move closer to levels 4 (High) and 5 (Full) automation in relation to cloud management activities.
Greater Levels of Abstraction – IT will continue to become more and more abstracted in 2020 and beyond (NoOps). The growth of serverless, containers, software-defined hardware, etc means that engineers / devs are thinking less and less about infrastructure. The focus away from operations and toward outcomes is another clear trend and likely one which will continue for some time.
Containers Become Mainstream – Application containerization is more than just a new buzz-word in cloud computing; it is changing the way in which resources are deployed into the cloud. More and more companies utilized containers in 2019 and we have seen estimates that suggest that one-third of hybrid cloud workloads will utilize containers in 2020 (ESG Research). Over the last couple of years, Kubernetes has established itself as the container orchestration platform of choice. 451 Research projects the market size of application container technologies to reach $4.3 billion by 2022 more businesses will view containers as a fundamental part of their IT strategy.
We have always enjoyed the quote ‘never make predictions, especially about the future.’ Nevertheless, entering a new year and a new decade it’s hard not to. We think the predictions above are fairly safe bets but equally, we are sure the speed and scale of change will likely be faster than we predicted.
With another recent round of earnings reports from Amazon, Microsoft and Google out of the way it’s always enjoyable to stand back and see what we can discern about the public cloud market share.
According to Synergy Research Group who closely monitor such trends, they saw 37% overall growth year-over-year in public cloud. They reported that it has taken just two years for the public IaaS and PaaS markets to double in size and their forecast shows them doubling within the next three years. Within the overall market it is possible to discern some interesting trends amongst the top three providers, which we discuss below.
Amazon Web Services (AWS)
Last Thursday, Amazon reported that its cloud division revenue increased 35% in the third quarter, which was down from 37% in the previous quarter, and its slowest growth rate in five years. AWS finished its third-quarter with $9 billion in revenue. Each of the three previous quarters also showed a decline in growth which can be seen below.
Microsoft followed AWS’s report with Azure reporting a revenue growth rate of 59%. In a similar vein to AWS, growth was reported as slowing and was down on the previous quarter which was 64% and down from 76% from a year ago. While Microsoft doesn’t break out specific revenue amounts for Azure (unlike AWS) Microsoft did report that its “Intelligent Cloud” business revenue increased 27% to $10.8 billion, with revenue from server products and cloud services increasing 30%
Azure also hit the headlines around the same time as their earnings report with the announcement of their securing the lucrative, high profile and highly contested $10B Pentagon’s JEDI Cloud contract. This was viewed as a key strategic win for the company and a game changer in the face-off with AWS.
Google Cloud Platform (GCP)
Last to report were Google’s parent company Alphabet. During their analysts call a few references were made to overall performance which included the Alphabet CEO calling out Thomas Kurian, who leads the GCP business, in saying “Obviously, ever since Thomas has come in, he has continued to invest across the board. He’s definitely focused a lot on scaling up our sales, partner and operational teams, and it’s playing out well”. Furthermore, it was reported that GCP had hired more sales, engineering and product managers, and that GCP, analytics and compute would continue to be a focus of the company’s investments going forward.
GCP falls into Alphabets “other” revenue bucket, which includes Google Play and hardware. Of businesses, GCP had the highest revenue. Other revenue was $6.43 billion in Q3, which was a 39% increase over $4.64 billion a year ago. There is no doubt that the cloud business is the largest of the three but Alphabet didn’t break out more specific numbers for cloud.
Some of the companies outside The Big Three, including Alicloud, IBM, Oracle, etc. are all growing, but they continue to lose ground to these three dominant market leaders. To compete, hyper-scale really matters, and these three bring that in spades.
Cloud in 2020 and beyond
As we enter the next decade a number of market watchers are speculating about what the reported slow down in growth means for the public cloud market share. As has been widely observed in other markets it’s a truism that as hyper-scale is achieved growth rates will decline. However, even with the overall reduction in growth they still exceed almost every other area within the broader technology market. As of Q3 2019, the overall quarterly run rate size was $25 billion, implying the annual run rate is now over $100 billion and still growing fast. It’s unlikely that there are too many other markets with better prospects going into next year.