Expect to see new uses for AIOps and observability as IT organization continue to adopt the concepts in 2022.
The year 2022 marks a time of maturation for AIOps and observability with an increasing focus on ways to use the evolving concepts.
Throughout the history of information technology, the adoption and implementation cycle for new tech has boiled down to three phases: If, When and How. Even Gartner’s Hype Cycle can be broken down into those three stages. Will organizations and users utilize the new tech? How soon will it be useful? And, what will the tech be used for?
True, there’s a fourth stage – Abandonment. But by time once productive technologies are ready for the trash heap it’s original implementors don’t care. They are retired or taking the eternal dirt nap. Even the PC is already in its fifth decade!
Few enterprises today aren’t using or at least planning for AIOps, and observability is proving itself to be a core element of successful AIOps and the overall business strategy.
So, what does 2022 have in store for AIOps and observability? The consensus seems to be that implementers will take the two to the next level as valued tools that keep systems running and supporting user and customer needs.
Here’s a sampling of tech observers and providers expect for the year ahead.
Hyperautomation and AIOps
In a recent presentation, Douglas Toombs, research vice president at Gartner, tied hyperautomation to the future for AIOps. “As hyperautomation is a critical path to achieve growth and operational excellence, I&O leaders must make automation a first-class discipline in everything they do,” Toombs said. That hyperautomation can be a key to effective AIOps and incident response automation, he said.
On the infrastructure and operations front — the systems that AIOps manages — Gartner predicts that organizations will turn to “Just-In-Time Infrastructure.”
“The speed at which infrastructure can be deployed is becoming just as important as putting the right infrastructure in the right place – colocation, data center, at the edge and more. This is the idea behind just-in-time infrastructure,” said Gartner analysts at the firm’s recent infrastructure and operations conference.
They added, “Borrowed from the term just-in-time manufacturing, this trend aims to reduce infrastructure deployment times as well as fuel enterprise responsiveness to business needs and anywhere operations. Gartner expects it to be a differentiating factor when enterprises compare and negotiate with service providers moving forward.”
Moogsoft Chief Evangelist Richard Whitehead outlined three key trends for 2022. He also cited Mordor Intelligence estimates that the AIOps market will grow from $13.5 billion in 2020 to $40.9 billion by 2026.
Whitehead’s predictions for 2022:
“Trend #1: enable remote and hybrid work”
Whitehead noted that the shift to work-from-home during the pandemic has complicated data collection for IT teams. “Businesses supporting remote work sent employees home with new hardware and software, resulting in more data traffic. And IT teams, already contending with increased data production, also had to monitor streams of data with different properties.”
He explained how an AIOps platform’s intelligent algorithms help IT teams handle that increased and increasingly dissimilar data. The system looks at the aggregated data to detect patterns and predict problems before they cause disruption.
“Trend #2: automating cybersecurity”
Whitehead said, “AIOps, traditionally used by IT operations teams, will also help enterprise security operations teams maintain constant vigilance of their systems. AIOps uses intelligent algorithms to model the systems’ standard behavior patterns and set baselines for system performance. These platforms unlock the ability to proactively detect a cyberattack by identifying deviations in real-time and determining if a performance issue is due to a cyberattack rather than another IT issue.”
When security issues do arise, the same AIOps platform can identify the resources or people needed to remediate the problem.
“Trend #3: decrease MTTR with observability”
Whitehead said the customer fallout from Facebook’s October outage is an example why faster mean time to repair is critical to keeping users happy and productive. “While technology can’t yet provide 100% protection against service failures, a platform that combines the alert data associated with AIOps and the telemetry data associated with observability can mitigate the damage. When issues occur and every moment counts, AIOps helps SRE and DevOps teams quickly detect the incident and provide actionable insights to help resolve it.”
See also: The Future of AIOps This Year and Beyond
The bots step up
In other predictions, authors at India’s Analytics Insight foresee new levels of virtualization and new uses for chatbots.
- Virtualized Operations and Service Management: Throughout the Covid-19 pandemic, the virtualized way of communication has been the new normal. AIOps solutions will ensure the smooth functioning of the business operations.
- Chat-bots: Chat-bots and virtual assistants provide automated support, reducing the need for live customer service agents. AIOps can be used to improve employee productivity as machines will provide them solutions to all their queries.
IT operations aren’t about just keeping machines running. It’s about making better use of the data generated by those machines and the people who use them. That’s not a new idea, but 2022 should see more organizations making that data a real tool of productivity.
TDWI predicts, “Machine learning enables predictive database indexing, analytics, and more.”
Organizations that have utilized machine learning and predictive analytics to anticipate issues such as network outages can apply those same concepts to the database itself.
According to TDWI, the complexity of the enterprise database has made it difficult for human administrators to manage storage and understand factors such as data usage patterns. “In contrast, ML-powered solutions can actually create data indexes, perform reindexing, and manage storage using predictive models able to guess where data sits.”
TDWI also predicted, “Where data and analytics vendors steer solutions away from open source licensing, communities will steer them back.”
TDWI said, “It’s crucial for organizations to remain aware of any licensing changes affecting solutions they rely upon for data technologies, as well as any open source options that become available to them due to this trend. If communities do prove their power to bring solutions back into the open source fold, it will discourage future licensing shenanigans and further ensure that valuable features remain available to everyone.”
James Connolly is a technology journalist with deep experience as a reporter, writer, and editor. He formerly was lead editor with Informa/UBM’s InformationWeek, Network Computing, and All Analytics media sites. He has more than 30 years experience in tech publishing, including work with Computerworld, TechTarget, and MassHighTech. He has covered a wide variety of information technology sectors with a focus on how enterprise organizations implement and derive benefits from tech. He is a former news reporter for the Boston Herald.
