Fotolia

Tip

IT automation software: Top five common mistakes

IT automation software projects can involve a great deal of risk, but avoid these common mistakes and it will increase the chances for automation success.

Automating data center processes may sound like a cure-all for the recession-fueled manpower gap in today's IT departments, but that isn't always the case, according to IT experts.

Sometimes companies find that they can't afford the initial investment required to make IT automation software work. Other times, experts say, companies embark on an IT automation plan, make a mistake or plan poorly, and end up getting sucked into a money pit.

"Automation is clearly something that is desirable because you can lower your costs by taking people out of the equation as much as possible," said Tony Iams, managing vice president at Gartner and formerly with Ideas International in Rye Brook, N.Y.

Anyone who is savvy appreciates the value that automation brings, but, at the same time, appreciates the huge risks.
Tony Iamsmanaging vice president, Gartner

"On the other hand, if you're applying automation, you'd better make sure it works correctly or you're setting yourself up for trouble," he said. "You don't need a lot of imagination to figure out what can go wrong."

Iams said companies that get the most out of IT automation projects tend to be those that proceeded with extreme caution when adopting and deploying the technology.

"Anyone who is savvy appreciates the value that automation brings, but, at the same time, appreciates the huge risks," he said.

1. Failure to communicate IT automation ideas

Companies looking at IT automation software often assume that they can get started with the technology right away, said David Williams, senior vice president of portfolio strategy at CA Technologies and formerly a research vice president with Gartner. The truth, however, is that their IT infrastructure often isn't set up to properly use automation. "This doesn't only apply to big enterprises; this applies to all companies," Williams said. "They're not always organized in a way that enables automation to be easily digestible."

IT infrastructures generally consist of specialists who focus on a particular area or discipline, such as server management, networking or applications management. But automation can easily bleed into multiple domain areas.

For example, Williams said, the networking infrastructure may need to be orchestrated with something in the server infrastructure -- and when you make that change, you may need to include some of the security components.

"When you automate something, it's nice to think that you can automate in a silo," Williams said. "But most automation value is across IT organizations."

2. Too much focus on infrastructure automation tech

It's easy to get caught up in all the hype surrounding popular technologies like IT automation software, but IT managers should really consider the business metrics they're trying to improve.

In other words, Iams said, don't set out to automate -- set out to solve a problem. Companies should identify the business problem they want to solve, and then try to solve it using the technologies that make the most sense. For example, a company that wants to reduce headcount might start with the goal of increasing the number of servers that each systems administrator can realistically manage.

"That would be the real goal, and if automation helps you get there, fine, but there may be other things that can help you get there," Iams said. "It could be provisioning software, for example, or it could be a number of other tactics that may help you increase the ratio of servers to administrator. Have metrics that make sense. Don't just focus on the tactic for the sake of focusing on the tactic."

3. Failure to understand the reasons for automating

Companies interested in IT automation software need to decide if they are automating primarily to save costs or to increase efficiencies, Williams said.

The answers to those questions should have a major impact on the automation implementation path the company takes.

Companies looking to drive out costs or reduce headcount would likely choose to automate many small, repetitive tasks, Williams said. Doing so may drive some of the minutia out of day-to-day operations and ultimately allow companies to decrease staff.

Firms that want to use automation to increase efficiencies could conceivably take the same approach, but they're better off automating big things, Williams said, like the addition or removal of virtual components and configuration changes.

"The full value of the process of automation is to get the big things done," the analyst explained. "Doing this means more skills would be required and more effort would be required. You're not going to drive costs out necessarily, but you are going to increase efficiencies."

4. Failure to build trust in automation technology over time

Smart users proceed cautiously with automation technologies, Iams said, and they have extraordinarily long testing processes for anything that brings automation into the equation.

"What that means is that after you set up some framework for automating tasks, before you actually allow it to take actions, you continuously use it on an advisory basis," Iams said. "You let that play out for a couple of weeks to a couple of months, and [when you're sure you trust it], let the system do it by itself and see what happens."

5. Not tracking the value of automation

IT automation software quickly becomes an unsung hero of the IT infrastructure when it's working properly, and folks tend to take it for granted. That's a big mistake, according to Williams.

Companies need to track the value of their automation software investment to make sure they're getting a return.

"You've got to make sure that the automation you're building has a reporting mechanism that continually reminds you of the value it's providing you," Williams said. "Then you can justify more automation -- you can justify more investment in the tools which allow you to automate."

IT automation software tools and companies

Product

What is it?

Features

BMC Atrium Orchestrator

A process automation system to create, coordinate and standardize automation workflows across applications, platforms and tools.

Uses prebuilt runbooks or allows users to develop custom runbooks.

Supports simple configuration of application interfaces through connectors and a mapping wizard.

Scalable architecture to accommodate large enterprise environments.

Improves service delivery, compliance and IT spending.

CA Process Automation

 

IT process automation tool that enforces standards across systems and organizations.

Design, deploy and manage the automation of IT operational procedures for the enterprise.

Automation and orchestration functions to increase IT staff efficiency, accelerate service creation and delivery, and reduce services costs.

Supports enterprise compliance policies and objectives.

Chef Automate

Configuration management and continuous delivery framework that enables infrastructure as code.

Creates and deploys code on infrastructure.

Common and customizable automated deployment patterns to package, configure and test code.

Ensures infrastructure remains in a desired state.

Collaborative platform for developers and operations staff.

Cisco UCS Director

 

Provisioning and IT lifecycle management software for compute, network, storage and virtualization resources.

Automates, orchestrates and manages Cisco and other data center hardware.

Replaces manual processes with automated workflows.

Assigns and configures resources for physical and bare-metal data centers.

Maintains isolation and security.

Supports Cisco UCS converged and hyper-converged infrastructure.

Real-time diagnostics, analytics and reporting.

Hewlett Packard Enterprise Data Center Automation

A lifecycle management product to provision and patch infrastructure, audit compliance and remediate issues for IT environments.

Orchestrates IT infrastructure lifecycle processes and automates tasks.

Enforces security and compliance rules.

An orchestration engine organizes processes.

Analytics available for resource capacity and performance planning.

Applies to virtual and physical servers, as well as databases and middleware.

Microsoft System Center

A management and automation system for Linux and Microsoft servers, cloud and on-premises infrastructure, as well as a range of compute, storage, network and security elements.

Uses heterogeneous and open systems, including Linux, Microsoft Hyper-V and VMware virtualization.

Deploys and manages Windows Server and Windows across the enterprise while monitoring system configuration, health and compliance.

Automated workflow processes and self-service options.

Helps diagnose and troubleshoot infrastructure, workload or application issues.

Puppet (Enterprise)

A framework and language for SysOps pros to define operations that support automated software deployment.

A standard way to provision and operate software in the infrastructure.

Orchestrates change with control, including designs provisioned across heterogeneous infrastructure.

Manages infrastructure as code for continuous software delivery.

Red Hat Ansible Tower

 

A management console for Red Hat Ansible that enables scalable IT automation and complex deployments with centralized control of the IT infrastructure.

Dashboard of real-time data showing Ansible host and inventory details and recent job activity.

Playbook display illustrates plays and completed tasks by machine along with results.

Workflows provision machines, apply base system configurations and deploy applications via playbooks.

REST API and command line interface can embed Tower into other tools and processes.

Salt/SaltStack

An orchestration, remote execution and configuration management system based on a dynamic infrastructure communication bus.

Controls cloud and container infrastructures, including multicloud orchestration.

Compatible with IBM, Amazon Web Services, Microsoft Azure, Google and OpenStack deployments as well as Docker and LXC containers.

Compliance enforcement and audits based on infrastructure security policies, with automated security remediation.

Continuous code integration and deployment.

Automatic configuration drift detection and remediation.

VMware vRealize Orchestrator

A tool to create and automate complex IT workflows within the vCenter Server platform.

An extensive workflow library with prebuilt tasks, or ability to create new tasks for workflows and other actions.

Enforces change control.

Operate thousands of simultaneous workflows.

Restart server without losing workflow states.

Centralized console for configuration, monitoring and troubleshooting.

Chart provided by Stephen J. Bigelow.

Next Steps

Make sure you get your IT automation strategy right

Infrastructure as code tools are adjusting to infrastructure changes

Adopting a DevOps workflow takes a process change -- embrace it

Dig Deeper on Systems automation and orchestration