Tierney - stock.adobe.com
Security and business agility have a lot to do with network automation adoption, but the various automation approaches available create a barrier for interested IT teams.
Organizations face pressure to transform their networks to keep up with customer demands and business innovation. IT teams transition to cloud, hybrid cloud, multi-cloud, edge computing and several other infrastructure designs that provide more scalable resources and faster data processing. But these designs also open the network to vulnerabilities, due to workloads and endpoints that span the expanded infrastructure.
Automation can act as a bridge between agility and security -- both of which are main drivers of network automation adoption, according to a recent report from Juniper Networks. In its "State of Network Automation Report," the vendor queried 400 networking and security professionals from U.S. enterprises and service providers to examine automation adoption, tools and challenges. Juniper found 96% of respondents had started implementing automation, with 67% citing security as the top technology driver and 60% citing business agility as the top business driver.
Why is network automation important?
Most traditional security measures can't keep up with the scale of modern network environments, but newer security products that tout automation and machine learning capabilities can be overly complicated. In response, many IT teams have started setting up smaller, realistic processes that automate blacklisting, vetting and firewall configurations. These processes then result in more reliable networks, according to the report.
State of Network Automation ReportJuniper Networks
"Automation enables reliability. Reliability leads to predictability, and predictability is an important part of modern information security," the report said.
As organizations implement these automation and security measures, they can also better prepare for strategic and transformative plans, instead of taking a more reactive firefighting approach, the report said. This can, in turn, boost network and business agility.
Network automation challenges and tools
While organizations are interested in network automation adoption, over half of respondents said they faced an overwhelming number of available choices during their research. Another factor that caused confusion is the discrepancy in how organizations label network automation.
While some IT teams deem software-defined networking (SDN) as automation, others define automation as using Ansible to push configurations. For example, respondents indicated they deployed automation in the following ways:
- 53% used configuration management tools, like Ansible and Chef;
- 50% used event-driven frameworks, such as Salt, StackStorm and SaltStack;
- 38% used custom monitoring tools and telemetry;
- 37% used source code management and infrastructure-as-code tools;
- 36% used container tools; and
- 35% used SDN tools with a centralized controller.
With so many flavors of network automation, it's little wonder why IT teams struggle to implement the best strategy for their network and applications. Most find it difficult to piece together a comprehensive approach that meets their requirements, especially when considering staff skill sets and resources, the report said.
For many teams, the first step is to automate network provisioning, which has long been lauded as a major automation use case. Configuration management tools that support deployment, testing and other provisioning factors are well established, are widely available and lead adoption rates among network automation tools, according to the report.
But IT teams should look at the whole picture, instead of focusing their automation efforts solely on provisioning and configuration management. This includes processes like troubleshooting and workflow and network monitoring, which respondents said take up large chunks of their daily responsibilities.
Ultimately, IT teams should try to create a "culture of continuous improvement" that enables trial and error, the report said. This type of environment supports strategic automation that teams incrementally integrate "into every aspect of daily operations" as they test, deploy and manage their infrastructure.