The Multi Cloud hype is increasing every day with both legacy vendors and public cloud providers (more specifically, Google which is lagging behind both AWS and Azure) talking about Multi Cloud, there is quite a bit of confusion on whether Multi Cloud will help avoid lock-in. A recent conversation on Twitter provided arguments on both sides of the debate. In this post, we will tackle the issue of lock-in in the context of multi cloud so that enterprise decision makers get some clarity on the right strategy for their organization.
The question of vendor lock-in
Vendor lock-in was expensive in the traditional software world. However, the economic model of cloud along with the continuous innovation by providers driven by the competition in the space has ensured that enterprise decision makers are not feeling the economic pinch of vendor lock-in. So, if you make a case for lock-in using the risks associated with getting locked into a single vendor, people are going to make a case that it is not hurting them as the cost of getting started with a new provider is not too high. At this point of enterprise cloud adoption, there is not going to be enough support for the risks associated with vendor lock-in.
Multi Cloud is about flexibility to innovate
Instead, one should frame the question of lock-in from the point of view of having the flexibility to innovate. The flexibility to innovate rapidly is more critical than the cost of getting locked-in to a single vendor. If your lock-in to the technology of single provider slows your organization down, you are letting a competitor or an entirely new player from an unrelated vertical to out-innovate your organization, potentially disrupting your entire business. The mantra for the Modern Enterprise decision makers should be
Retain the flexibility to innovate rapidly or get disrupted
The threat of disruption through rapid innovation in technology is what is keeping innovative CIOs awake in the night. By losing your flexibility to change course or build a new feature or a service in a short timeframe is critical for any modern enterprise. Multi Cloud becomes an important part of your strategy if you approach lock-in from this point of view.
Every single cloud provider has a different set of services. This impacts the architecture, dependencies, etc. for the applications. When you are locked into specific set of services from a single provider, your ability to take advantage of services offered by another provider (say Google ML capabilities) is not going to be easy. Yes, there are efforts like Open Service Broker API gaining momentum but it is a long way to go before ISVs across the industry embrace the API or all the cloud providers treat them as first class citizens. The market for modern IT stacks is still wild west. Cloud providers and ISVs are marketing themselves in all possible ways including partnerships with everybody and anybody. Soon, the market is going to consolidate and business reality will come to bite these cloud providers. We will soon realize that all the support for OSS foundations and partnerships with competitors are just pure marketing with very little value to end users.
This puts the onus on the enterprises to have a strategy that gives them the flexibility to innovate as technology changes rapidly and as they feel the heat from their own business realities. The only way to retain the flexibility to avoid lock-in to specific technologies provided by a single provider, the new central dogma for the modern enterprise.
At StackSense, we are committed to sharing our knowledge about modern stacks to help enterprise decision makers in their modern enterprise strategy. We will be posting a series of posts related to multi cloud that will help in the process.