The promise of an automated solution portal is quite appealing to CIOs these days. Their IT staff is already burdened with a series of checklists and process controls needed to ensure the company’s solutions are delivered correctly and securely. Having a portal to handle even some of this overhead is a welcome relief, but it’s not enough to just make a decision to start automating process. The move to an automation solution is a strategic shift in the way services are consumed, and it should be planned for accordingly.
Sometimes planning what not to do can be just as beneficial as figuring out what to do. In my day to day as a Senior Data Center Engineer, I see clients fall victim to issues that come up even before the first piece of software is deployed. In this post, I’ve listed three of the most common you should look out for in your own automation project.
Pitfall #1 – Expecting to automate everything quickly
Organizations sometimes jump into an automation project with both feet and a desire to reinvent their strategy. It’s great to start with this type of enthusiasm, but be careful that you don’t try to run before you walk as a typical automation project is usually slow to show gains.
One of the first automation tasks usually targeted is the building of a virtual machine. Before you tackle this step, stop and think about some of the activities that need to be in place before you can do this in an automated fashion:
- Get approval for a new virtual machine
- Get an available IP Address
- Enter information into a Change Management solution
- Update a monitoring tool
- Deploy anti-virus software
- Send out an email to let users know their machine is available
The list above doesn’t even include getting your automation portal deployed, orchestration configured, or any of the common installation requirements for an automation solution.
If you’ve decided on an “Automate Everything” strategy, then you should be aware that many of the gains you’ll see will be months away from the initial onset of the project. Set your expectations that this is a long-term goal and start with some of the smaller tasks that will yield the most benefit.
Pitfall #2 – Not getting participation from all teams
Sometimes a new “Cloud Team” might be assembled whose only focus is building the organization’s automation and orchestration solution. While it’s great to have dedicated time to work on this project, it may be more beneficial to have members from each of the infrastructure teams (Network, Storage, Compute, Virtualization) working on the project. Members from each discipline should still belong to their respective groups but also belong to a team that has an “automate first” mindset.
With each infrastructure team providing feedback to the Cloud group members through their own liaison, they can feel like part of the process instead of getting the notation that the automation system is replacing their job roles.
Pitfall #3 – Not focusing on standardization
When planning to automate processes, they need to be run the same way each time. If the organization can’t settle on a standard process to deploy VMs or introduce new development code into the organization, the automation piece becomes much less effective. This sounds like a pretty simple thing to do, but you may be surprised to find out that some groups just want to deploy stuff without a standard, and an automation process may be perceived as a roadblock to this type of structure.
Decide early on what the strategy is so that the automation solution isn’t just another process to be managed. The goal with automation is to replace the old processes, not to add new ones that also need to be maintained.
At AHEAD, we help IT teams recognize and overcome these obstacles every day. If you’d like to discuss your own automation project, please schedule some time with us to chat about your challenges.