This year will see a marked increase in competition for public cloud dollars – and not just from incumbents like Amazon and Google. As demand for the cloud grows, big companies from China are making significant moves to expand their global reach – with Alibaba in particular moving aggressively into Europe and the United States.
The public cloud spend of a typical Fortune 500 can quickly escalate to eight or nine digits on an annual basis. Snap alone is paying hundreds of millions of dollars a year on Google Cloud.
Proprietary services and software that are part of a public cloud’s PaaS offering create significant lock-in – and credible, compatible alternatives often don’t exist
Winning one of these big customers means big money and the competition for marquee contracts will just be heating up this year. And with so much competition, there is going to be opportunities for huge savings – for companies of all sizes.
While the biggest winners will be those enterprises that have invested heavily in the cloud, IT organizations need to make sure that they implement the right cloud infrastructure and technologies to make the most of lowering costs.
One key solution is multi-cloud.
Vendor lock-in will be cloud issue #1 for the enterprise
No decision maker worth their salt is going to want to embrace a single cloud platform to the exclusion of all others. While some organizations might lock in a really good contract, it won’t be with any of the big 3 unless they cut prices dramatically. But these forced price reductions will only further the cycle of cost cutting and drive enterprises to keep their options open.
As a result, the level of anxiety for decision makers at big organizations around vendor lock-in will continue to rise – already vendor lock-in has replaced security as the #1 cloud concern.
While most organizations will remain on a single cloud this year, they will be actively seeking out options to avoid being trapped on that cloud. More and more enterprises will be moving away from expensive and limiting proprietary cloud storage technologies developed by Amazon, Google and Microsoft and embracing open source software solutions. As a result, it will be the beginning of a bad set of years for the highest price, stickiest services that are being offered by cloud service providers.
What’s at risk for enterprises considering proprietary solutions? Take Snap – a company that uses App Engine – a Google platform. App Engine is 10x more expensive for Snap than other solutions but they are stuck on the platform because to move away from it would mean rebuilding. This would require a huge investment in engineering resources and the potential of mass instability of their platform – which could drive away users. You’ll see fewer and fewer enterprises falling into that trap in the future which is why we are seeing so much interest in multi-cloud strategies. Conversely, it will be a good year for tools that make it easier for enterprises to avoid vendor lock in – like Kubernetes and Docker.
The best path forward to multi-cloud
Most large enterprises have instituted mandates for a multi-cloud strategy. SMBs would do well to plan for one, even if it will not be a practical reality in the near term. There is a non-trivial cost to building for multi-cloud deployments, because they require a layer of abstraction between a company's IT footprint and the underlying cloud vendor's APIs. However, not building from the start for multi-cloud makes the eventual transition increasingly difficult, as each additional vendor-specific hook is utilized directly by deployed services. The good news is that the necessary layer of abstraction is rapidly evolving via open source and commercially-supported offerings, including Kubernetes, Docker, DC/OS, and Cloud Foundry.
Cloud price wars will heat up this year – it will be the beginning of a bad set of years for the highest price, stickiest services that are being offered by cloud service providers
Implementing a multi-cloud strategy requires first and foremost that IT leaders select cloud-neutral technologies. Proprietary services and software that are part of a public cloud's PaaS offering create significant lock-in. Credible, compatible alternatives often don't exist, and worse, migration paths are incomplete and poorly supported; it's not in a cloud vendor's interests to provide an easy off ramp. For two arresting examples, look no further than Dropbox's struggles to replace their usage of AWS S3 or the aforementioned Snap and their ongoing battle with Google App Engine's explosive cost structure.
Cloud price wars will heat up this year. Putting all of an enterprise's eggs into a single cloud vendor's basket invites risks resulting from the vendor's potential systemic security and/or operational shortcomings. A multi-cloud strategy will give enterprises the flexibility to migrate between cloud vendors – and take advantage of falling prices.