Скачать или смотреть What Is Vendor Lock-In? | Dirk Van de Poel, CPO at Klarrio

We all agree that vendor lock-in is undesired, but how can you actually avoid it?

There is a lot of content on social media about why it’s crucial to avoid a vendor lock-in situation. Still, we see fewer discussions about how to actually achieve this.

Here are some of the more technical aspects of how we at Klarrio help organizations avoid vendor lock-in:

🔹 Assess your organization’s specific needs: We always start by having a conversation with our customers to find out pain points and where they are exposed to vulnerabilities.

We know this might be surprising, but vendor lock-in is not necessarily the worst-case scenario for every company*. Every time you face a technology decision that may lead to lock-in, you have to weigh the cost of switching against the cost savings offered by the lock-in alternative. This cost-benefit trade-off differs for each organization and depends on company size, in-house talent, the amount and business-criticality of the data involved, and other factors.

🔹 Design with portability and resilience in mind: If you rely only on proprietary services, you cannot migrate without a significant effort involving rewriting queries and possibly re-architecting apps. That’s why we recommend using open-source software as a building block, as it allows you to run your solution in different environments.

Using containers and orchestrating them with Kubernetes can avoid tying apps to a single cloud’s proprietary services. It’s also helpful to keep infrastructure defined as code so it can be deployed to multiple providers.

In addition, using standard APIs and service abstraction will ensure your applications don’t depend on one cloud. Pairing this with vendor-agnostic CI/CD tools will allow your build and deployment pipelines to target any environment.

🔹 Use abstraction layers in the infrastructure: Decoupling your applications from vendor-specific services is essential in avoiding vendor lock-in.

We suggest opting for multi-cloud–compatible databases (such as PostgreSQL or MongoDB) and using open standards for messaging and streaming (e.g., Apache Kafka) instead of proprietary services.

🔹 Devise a portable data strategy: It’s crucial to store business-critical data where you can access it with different technologies whenever you want to. You can ensure this by storing your data in open formats and replicating critical datasets across multiple clouds or an on-premise server.

We recommend assessing your organization’s specific needs. For example, object storage and open table formats may be suitable for large data sets, while in other cases, we advise going for PostgreSQL or one of its many variants.

🔹 Make sure you have an exit strategy: Having a clear plan for migrating your applications and data away from a provider will enable you to avoid being trapped in a single ecosystem, facing massive re-engineering costs if you ever need to switch.

Find out more about how we help organizations: https://klarrio.com/manifesto/

#Klarrio #StreamingAhead #SecurityByDesign #CloudAgnostic #Vendorlockin #Cloud #cloudcomputing