Intelligent building technologies are wonderful -- as long as they’re operating properly.
As building owners continue their foray into various IoT systems that are being deployed across commercial buildings and campuses, managing hardware and software of each system becomes a tall order if not properly handled from both a performance and cybersecurity perspective.
Fortunately, advanced monitoring platforms that take advantage of artificial intelligence (AI) are becoming widely available on the market. They’re known as artificial intelligence for IT operations (AIOps) platforms -- and they may be the exact solution for your smart building monitoring and management needs.
What is an AIOps platform?
Infrastructure monitoring and management tools have been around for decades. Examples include simple test tools that use ICMP and SNMP protocols all the way to advanced systems that perform deep packet inspection (DPI) and streaming network telemetry of every data packet that crosses the wire.
While all these monitoring protocols, tools and platforms provide monitoring visibility, most are deployed independently, offer no automated analysis and do not share critical data between each other. An AIOps platform consolidates many of these disparate technologies into a single unified tool that is centrally managed.
The concept behind AIOps is to collect large amounts of relevant network-based monitoring and health information. So much so that human administrators would find it difficult, if not impossible to manage. However, with the help of AI, the data analysis process can be performed automatically with alerts and remediation steps being presented to the operations administrators with relatively little manual intervention being required.
How does an AIOps platform gain intelligence?
To gain the necessary intelligence required to identify performance and/or cybersecurity issues, newly implemented AIOps systems must first sit on the network and observe what’s considered to be “normal” network behavior. Once this is accomplished, monitoring thresholds for various performance and security-related metrics can be set by administrators that will in turn alert them when network behavior veers outside those set limits.
Because the AIOps platform is ingesting and analyzing network health information across the entire network, AI can often determine which devices caused the issue and why in real-time. This significantly reduces the amount of troubleshooting time spent as well as the speed at which administrators can remediate a performance or cybersecurity incident.
How can AIOps help me from a smart building monitoring and management perspective?
It’s best to think of AIOps platforms as an AI-based extension of your IT operations team. Given this context an AIOps platform automates many processes that were previously performed manually by staff. This includes:
1. Collecting network performance, health and security data across a building or campus network.
2. Intelligent analysis of collected data from a performance and cybersecurity perspective.
3. Providing automated alerting when performance/security issues are identified.
4. Offering a centralized, end-to-end view of the overall health of the network and connected devices.
For building owners and operators that are seeking to integrate smart building technologies into their properties -- but who are also wanting to work under a "Lean-IT model" that limits the number of IT staff required to manage smart building technologies, AIOps provides the following benefits:
- Automated identification of IT performance or cybersecurity issues.
- AI-backed root cause analysis and remediation.
- An extra layer of cybersecurity monitoring to identify when compromised devices begin communicating with command-and-control servers, veering from “normal” network behavior.
Are there any drawbacks to AIOps platforms?
One significant drawback to an AIOps platform resides in the fact that the monitoring and alerting tool must first observe the infrastructure as it operates under existing conditions. This step is used to baseline what is considered to be optimal network behavior.
The baselining step can be challenging if devices or services on your network are already causing performance or cybersecurity issues. If issues are not resolved prior to the baselining process, the AIOps platform will build them into baseline metrics and consider them typical behavior.
This baselining requirement means that time and effort will have to be spent to manually resolve as many performance and security related issues ahead of time so that the AIOps tool will establish a monitoring baseline that’s a close to optimal as possible.
Thus, the best advice to those that are looking into AIOps to streamline their infrastructure monitoring and management is to first get all existing technologies operating as well as they can. Once that step is complete, ongoing AIOps monitoring and alerting will be far more accurate.
