Network discovery and mapping company NetBrain has released an automated network troubleshooting tool called the Problem Diagnosis Automation System.
NetBrain has positioned the PDAS as a more transparent alternative to AIOps in network management automation. Like AIOps tools, PDAS learns over time but maintains more visibility into how the network problem is resolved.
When PDAS encounters a network connectivity issue, it attempts to match the situation to matters met previously, including those discovered by third-party tools like ServiceNow. If it finds a match, it will automatically diagnose the issue and populate its findings directly into the IT service management ticket. NetBrain claims its approach can resolve up to 95% of all network service tickets, which tend to be recurring issues and prevent up to 50% of abnormal network conditions that would lead to outages.
When a novel network issue arises, PDAS can provide some information about what the problem isn't. Then, it will capture the steps a network manager takes to resolve the issue to make that resolution repeatable and shareable through an automation template. The templates let machine behavior trigger corrections rather than a network manager.
PDAS aggregates all the resolutions fed into the system, which means the product will continue to improve over time as it encounters more network issues. Unlike AIOps tools that learn from problems via AI, PDAS automates new processes from the network resolutions created by system engineers.
NetBrain's approach might prove to be its primary draw for some enterprises, said Shamus McGillicuddy, an analyst at Enterprise Management Associates. Many network managers who have been slow to adopt automation because they don't yet trust AI with critical network functions may feel differently about more transparent network automation like PDAS.
NetBrain approaches network problems from an intent-based standpoint rather than a device-based one, meaning the company's products focus on whether networking goals are being reached rather than whether individual network devices are functioning. In today's complex networks, devices can be fully functional as the communications between the hardware break.
PDAS is likely to see competition from intent-based networking, network modeling and verification, and AIOps vendors like VMWare's Veriflow or IP Fabric. But the technology most directly competing with PDAS are homegrown scripts that engineers write to automate recurring problems in their networks.
"This is [NetBrain] automating the whole process, where [PDAS] gets the tickets and analyzes the tickets and provides a more productized environment for you to build out the automation," McGillicuddy said. "I've talked to people in the past who say we see the potential of network automation, but it's not easy to implement -- this is them making it easier to implement."
McGillicuddy's research found that 89% of organizations find it at least somewhat important to have a data center networking automation tool with integrated monitoring and troubleshooting capabilities. But most network automation products of the last few years have focused more on configuration than troubleshooting, so NetBrain's approach could spark interest among some enterprises.
PDAS is available now. NetBrain has not released pricing but said the pricing model includes an upfront payment and a usage subscription.
Madelaine Millar is a news writer covering network technology at TechTarget. She has previously written about science and technology for MIT's Lincoln Laboratory and the Khoury College of Computer Science, as well as covering community news for Boston Globe Media.