Asteris

Climbing the DevOps Learning Cliff with Mantl

The learning curve to understand all the moving parts is steep. Some days I believe […] it’s almost vertical.

Ben Hall

Imagine this: you’re responsible for deploying an application for your team. Part of it is written in Java, part in Go, and part (somehow) in Perl. You go to the AWS console and stare into the abyss of icons. It looks back at you. So… what’s first?

You might start off spinning up some VMs, but what do you use for secret storage? And how about an overlay network? HTTP ingress? Oh, and you need a persistent disk for the Java portion of the app. What happens when the inevitable “we’re shutting down your instance in an hour” email arrives? Is it time to dive into the API documentation?

You’re certainly capable of learning all this, but do you have the time?

But what if the learning curve weren’t so steep? You could focus on developing your app instead of playing life support. The time you freed up would improve the business’ bottom line and you’d end up looking great.

It’s a nice daydream, but in the meantime… you’re stuck babysitting servers that halt and catch fire if even the smallest things go wrong.

The future of DevOps is ever-changing. With developers continuing to take on the responsibilities of DevOps, this requires tools that meet the needs of entire development teams, rather than a select few.

Kiran "CK" Oliver, The New Stack

Now take that line of thought, and apply it across your entire organization. Tens of teams deploying hundreds of servers for thousands of microservices. And each one of them doing their own operations in their own way, never talking to each other. Yikes, right?

It’d be great if people would, you know, talk to each other about these things, but they don’t.

Shared configuration could be great. Teams would stop siloing their knowledge and share it across the organization.

But, real talk: how do we get there?

Situations like these play out across thousands of businesses just like yours, every day. People are spending all their time reinventing the wheel!

This was the problem that Cisco brought to Asteris.

[Mantl] addresses this need and meets the challenge of orchestrating a microservices manager for DevOps teams working with both enterprise and small-scale applications.

Kiran "CK" Oliver, The New Stack

From this problem, Mantl emerged. Mantl deploys a full container stack from servers to load balancers in minutes, allowing teams to stop reinventing the wheels that their businesses run on. We combined top open-source tools like Kubernetes, Calico, Consul, and Terraform to create an out-of-the-box experience that can run on any cloud from AWS to vSphere.

We worked with Cisco and the community to create a system that would suit their needs. Today, Mantl is suitable for such diverse needs as application deployment and data analytics, all in the same cluster. Mantl is one of Cisco’s most successful open-source projects, with over 400 GitHub forks and 2,500 stars. New tools have come out of our development partnership with Cisco, such as mesos-consul, consul-cli, and terraform.py.

When teams use Mantl, everybody’s deployments stay on the same page. There are no more “snowflake” servers, everything is commoditized. Cisco has been able to use this to coordinate internal teams around the same tech stack. When a new team wants to spin up cloud services, they use Mantl as a common language.

When you use Mantl and that team comes to you with their hodge-podge of an application, you’ll know how to deploy it quickly and scale up.

Get Your Organization on the Same Page

Asteris specializes in DevOps tools. We can help you get your organization on the same page, and tame the learning curve. Contact us to make your DevOps practice shine.