Infrastructure services made easy: 4 essentials for transparent IT Ops

Your developers are using open-source tools, skirting compliance and resource consumption practices that your IT operations management organization has put in place. So how do you get them on board?

Transparent operations is a way to ensure compliance with internal standards and cost controls—two critical capabilities for any organization moving toward an integrated DevOps methodology—by monitoring and tracking the infrastructure services that your development team uses. By implementing transparent operations, you'll be building a better partnership with your development teams, and they will be more likely to consume infrastructure services from a catalog that the IT operations team hosts.

There are four essential requirements your products and services must support when implementing a transparent operations model:

  • Organization-wide governance
  • Compliance with security and privacy requirements
  • Cloud-bursting transparency
  • Flexibility to choose the level of abstraction for application hosting

Here's a closer look at each of these requirements.

Software Defined Networks Have Arrived: Why SDN Needs Management Tools 

Organization-wide governance

Your transparent IT operations management system should allow for governance across every environment available to development teams at each stage of the application release pipeline—and for good reason.

Say IT Ops wants to run a production application in the cloud. That doesn't mean you should also deploy a testing version of the application in the cloud. Testing on cloud infrastructure can create bottlenecks, because load tests consume a huge amount of bandwidth and CPU resources. Instead, set a policy across the organization to ensure that 1) production applications are approved to run on cloud, and 2) all applications in QA, performance testing, or the development stage use your local virtual machine implementation.

Your system should facilitate policy-based governance and automate the switchover to the appropriate environment at each stage of the release pipeline. And for the development team, proper support from infrastructure should be completely transparent.

Compliance with internal security and privacy controls

Developers don't want to worry about infrastructure security. After all, they have their own app sec practices to worry about. When they're deploying an application in any environment, the security of the underlying stack should be a given—that is, transparent. They shouldn't have to worry about compliance, or whether the app server, the database server, or the operating system layer has security issues or has been certified by IT Ops.

The IT Ops team's transparent operations capability should ensure that applications are always deployed on a secure, highly available environment that's been certified by IT operations. And every item in the IT operations service catalog should offer the required security and scalability.

Support for transparent cloud bursting

Developers often need to scale outside of their infrastructure. For example, you might have a private cloud infrastructure that can handle up to 100 instances of an application, and you might have scaled up to the point where your private infrastructure is fully consumed.

Now you need to scale outside of your private domain. Developers should not have to worry about how to do that, or about where their applications are hosted at any given time. IT operations should guarantee that cloud bursting happens in a scheduled, repeatable, and potentially automated way, without developers needing to know what is going on.

Organizations need to use their infrastructure to the fullest before expanding to the cloud. But your development team may not be aware of the infrastructure resources available to them internally, so when they buy directly from a cloud service provider, existing infrastructure investments might be underutilized.

Flexibility over specified levels of abstraction

If you look at the evolution of IT infrastructure, abstraction increases as you rise through each of the supporting layers. Today, almost everything is available as a service.

  • With infrastructure as a service (IaaS), the physical infrastructure was abstracted.
  • With platform as a service (PaaS), it was not just infrastructure but the underlying platform that gained abstraction; you didn't care which middleware or OS you were using.
  • With container as a service (CaaS), all the container engines, orchestration, and other resources are provided to you.
  • As the industry moves to serverless computing, everything is abstracted except the application itself. You simply write an app, zip it, and upload it to, say, AWS Lambda. You don't worry about the app server, the database server, or the rest of the stack.

Because you always have different teams working at different layers of the application pipeline, a good, transparent IT operations management product should give you the ability to choose the level of abstraction you need. For a new application, you might want it to run in a serverless environment. But when moving an existing, on-premises application to the cloud, you might prefer to use IaaS or PaaS.

Different use cases require different levels of abstraction, and your transparent operations solution should give developers the flexibility to choose the appropriate level.

Make your IT Ops truly transparent

If your IT Ops team insists that developers write for a specific app server or database server, or that they must use a specific stack, the developers will simply bypass the IT Ops team. And making it easy for the development side of the house to work more closely, and in compliance with, the operations side of the house is one of the main reasons to adopt a transparent operations capability.

Simply put, enabling transparent IT operations can help engender the tight relationship between operations and development that's required for integrated DevOps.

Software Defined Networks Have Arrived: Why SDN Needs Management Tools 

Has this article been helpful to you? Is your IT Ops team already on its way to implementing easier methods and infrastructure for your development teams to consume? Share your thoughts and experiences in the comments section below.

Topics: IT Ops