What is Vendor Lock-in and How to Avoid it
A Brief History of Data Storage
In the past when Unix mid-range systems dominated enterprise data centers and computer rooms, computer manufacturers controlled a significant portion of the hardware and software stack, from the proprietary CPU to the operating system to the application environment and all the way to the staff needed for support.
For example, moving from DEC (Digital Equipment Corp) to Sun Microsystems in the server room was a significant endeavor that required extensive planning and resources. However, once intel-based servers emerged, the game changed, and contemporary server manufactures rarely have the clout to monopolize a company’s server room any longer.
Today, it’s common to see enterprises use multiple manufacturers at the same time as servers become more and more commoditized.
Why is this Important?
Because the winner here is the customer. More choices and low barriers to switching mean vendors have to earn every dollar they make from you.
Lately, we’re seeing a similar evolution away from vendor lock-in and a move toward low barriers to switching happening in the cloud Infrastructure as a Service (IaaS) market. In the early days of managed hosting and dedicated servers, you could lease a physical server with an operating system running on it, which is similar to what you can do today.
However, because server virtualization was not widely available yet, each physical server supported a single OS instance and backup/restores had to contend with OS drivers tied to the hardware of that machine. Moving an OS instance from one machine to another was possible but sometimes very hard, making switching costs fairly high. The rise of virtualization addressed this issue in the enterprise data center as companies learned to decouple server instances from the hardware on which they were running.
Navigating the Public Cloud
Recently, public clouds then came along and the VM displaced the physical server as the primary unit of computer consumption.
But with the greater usage of public clouds, a new vendor lock-in has emerged: the cloud service lock-in.
Public cloud providers make it very easy to migrate your infrastructure to their cloud, but they don’t really tell you how to get your resources out of it. All the while managing the cost of cloud consumption has become a full-time role in and of itself, and companies often feel like they don’t have many options.
Fortunately, like the commoditization of computer hardware before it, the evolution of the cloud is also moving in the direction of more customer choices and lower switching costs.
One technology that is fueling this evolution is the OS-level virtual container, which displaced the VM as the primary unit of computer resource consumption and organization.
Coupled with a hybrid cloud posture, the containerization of workloads can bring an unprecedented level of control to customers for how resources are deployed across a diverse range of infrastructure options while giving users the freedom to move resources between public cloud, private cloud, bare-metal, and on-prem environments with near-zero switching costs.
Read More from this Author