Code

No-Ops – Next Evolution or Fuss?

Will the so-called “DevOps” be replaced with another buzz word, NoOps? So, what exactly is NoOps, and will this be the next evolution or just an exaggerated practice. Let’s find out.

What is NoOps

NoOps means No Operations, i.e., remove all operations tasks and reduce the resistance between Development (Dev) and Infrastructure (Ops) teams. The primary goal of NoOps is to build a process where everything can be deployed in an automated fashion by Developers from the infrastructure to the application.

Origin of NoOps

With the adoption of Cloud technology, we can build complex solutions and solve numerous challenges. It has given rise to PAAS (Platform-as-a-Service) and Serverless services by Cloud Vendors like (Amazon Fargate, AWS Lambda, Azure Functions, etc.) whose sole purpose is to focus on the application code and leave the operational headache with the service provider. All of this contributes to removing the traditional operations activity, i.e., build and maintain (monitor, backup, patching, etc.) the infrastructure.

Opinions against NoOps

There are multiple reasons which are a hindrance to the wide adoption of NoOps.

  • Legacy Applications: In fact, every organization is running some legacy application either on Physical Server or on a Virtual Machine (Datacenter), and these applications cannot be migrated to Containers or broken down into microservices. They require operations team support.
  • Compliance and Security: If we want to have better control over security and compliance standards, we would require a skilled operations team. We cannot rely on the service provider for these things. 
  • Developers are not leaning to administer: I know this cannot be considered a benchmark in adoption against NoOps. Administrators are learning to code depending upon the requirement, and developers aren’t always interested in administering their solutions.    
  • Incident/Problem Management requires a fair amount of skills: Apart from harmful code, there are multiple reasons for things going wrong in a large environment. Isolating and recovering from the problem requires in-depth knowledge of underlying infrastructure, which is impossible without the operations team.

In the end, there is one conclusion. Neither NoOps nor DevOps is perfect. We all are evolving, and with rising technologies, we need to see what best fits our needs.

Leave a Reply