Five Things To Sell Your Boss On Container Automation

I was sitting on a plane waiting to take off for home after a day of speaking with VCs (a topic for another blog for sure!), when the fellow sitting next to me asked me what I did. I explained that I developed software that helped companies deliver and manage their applications more efficiently and he was interested because, as it turns out, he worked in IT Operations at a SaaS software company. As I got into the details of our solution and what it could do for him, he said, "Yeah, I get it! I really see the value for me and the rest of the team, but how do I sell it to my boss? I've gone to him before with technology that could provide efficiencies and speed time to market, but it seems he's only interested in the bottom line."

This really got me thinking about what is really important to a Director or VP of IT Operations in a company like my seatmate's. And really, it comes down to two things:

  • How can I reduce costs while maintaining or increasing service
  • How can I avoid hassles with the rest of the organization

So these are the five things I told this fellow to tell his boss to sell him on Container Automation:

  1. Bottom Line - He'll save 30-50% on his cloud infrastructure costs - With the appLariat Container Automation Platform you can use policy control to start/stop application deployments and scale up/down the size of your deployment cluster automatically. You can also set lease termination policies on a workload basis so that orphaned application deployments that are no longer being used won't take up precious cloud resources. You can scale up/down large production and staging application deployments based on load so that you don't have to have reserved cloud compute instances for your peak load. This means you only pay for cloud infrastructure when you are actually using it. As an example, one of our customers was able to go from 129 AWS 100% reserved instances to 65  50% reserved/50% hourly instances for a cost savings of approximately $15K per month. For this customer, the cost of our solution was $50K/year, so it paid for itself in just over three months. If that doesn't peak your bosses interest, I don't know what will!
  2. Bottom Line - He'll save 25-30% on DevOps/Admin support as the company and IT environment scales - Because cloud infrastructure management and application deployments can be done more efficiently, the existing DevOps and Infrastructure Admin team will have 25-30% more capacity. Your boss won't need to hire an army of DevOps engineers modifying and managing Chef, Puppet or Ansible scripts to deploy application versions for Development, QA, Staging, etc. as your company and offerings scale. And if one of his DevOps guys walks out the door with all the knowledge of the quagmire of deployment scripting you're reliant on now, it's a big problem. With appLariat, no scripting is required and updates to components, artifacts and configuration data are easy, with complete traceability.
  3. Avoid Hassles - He'll be able to provide production-like environments so that the Dev team can reproduce production bugs in a test environment - There is nothing your boss hates to hear more from the development team than, "We can't reproduce the bug you are seeing on the production site in the staging environment, so we are going to have to debug it on the live system." But it happens more than he'd care to admit and it's nerve-wracking. It's usually because the staging environment and production have diverged and no one knows exactly what is deployed in production vs. staging because the information only exists in a maze of deployment scripts. With appLariat you can deploy the current production release for testing at the push of a button and always have complete traceability to exactly what components, configuration and artifacts are currently running in production.
  4. Avoid Hassles - He'll be able to do component upgrades 50-75% faster so Dev can use the latest and greatest - Development always like to use the latest and greatest. But if you've got 27 applications in production that are all reliant on a particular version of MongoDB, you've got a serious component upgrade exercise to move up to a newer version that Development wants to use. Now your boss knows that this is not high on his priority list, but he's got Dev managers hassling him to do it so they can upgrade and take advantages of new features. This means his team is going to have to deploy these 27 applications one by one with the new Mongo version, run automated testing and record the bugs so that Development can fix them. This is exactly the situation one of our customers found themselves in - in fact, they had multiple components across the 27 apps that were now out of support by the vendor! With appLariat's fast, automated deployments they were able to realize a 50-75% reduction in estimated project completion.
  5. Avoid Hassles - He'll be able to enforce the production SOE throughout the development lifecycle so that apps don't fail when they reach production - You boss also knows that when he puts a new app in production it's likely to fail even though the development team says it works fine. Then the blame game begins between Dev and IT Ops. It's a total nightmare and takes forever to resolve. Meanwhile the business sees they are losing first mover advantage and is angry. A major analyst says that this happens 86% of the time and the main reason is that the production Standard Operating Environment (SOE) is not enforced throughout the development lifecycle. Your boss knows how it goes. Developers are using one version of Redis for unit testing, QA is using a different version and none of these is the currently approved version for production. Or no testing is ever done with monitoring or intrusion detection agents used in production. With appLariat, the production SOE can be defined and enforced throughout the development lifecycle, so your boss will know that new apps he needs to put in production have been tested with it. That means your boss can avoid one of his most painful hassles. And, of course, if the app fails, it won't be his fault.

My seatmate felt like he had some great ammunition to take back to his boss. I'm looking forward to hearing from him...

Close Menu