Early AIOps Adopters Not Sold on IT Advice
AIOps, the emerging IT operations technology that uses machine learning and AI techniques to automate digital infrastructure, has matured to the point where vendors are releasing surveys promising insights into how the technology is being implemented.
Among them is OpsRamp, the AIOps platform vendor, which released a survey this week that seeks to gauge how the ML-driven automation technology is streamlining IT operations and boosting infrastructure performance.
While the majority of the 200 IT managers surveyed said AIOps is mostly delivering on the promise of automating repetitive tasks, quickly analyzing incidents and generally boosting resilience and availability, issues based on user experience are also surfacing.
They include doubts about the reliability of recommendations delivered by AIOps tools, which could stem from a lack of data used to train machine learning models used in AIOps. Hence, just over half of those polled expressed “major apprehensions” about implementing AIOps tools. Lack of data science and machine learning skills also were cited as lingering concerns.
Nevertheless, startups like OpsRamp are betting that embattled IT teams will welcome automation tools as they scramble to harness hybrid enterprise deployments characterized by multiple clouds and cloud-native DevOps running in parallel with legacy infrastructure and workloads.
“AIOps is emerging as a real-world solution to the data overload, infrastructure complexity and incident remediation problems that are overwhelming digital operations teams in today’s enterprises,” Bhanu Singh, OpsRamp’s senior vice president of product development and cloud operations, noted in releasing the AIOps survey on Wednesday (May 1).
The startup’s value proposition is helping companies and service providers determine the “right levels of automation” needed to keep their servers running or helping to fulfill service-level agreements.
Incident management is critical. The AIOps survey found that “establishing data accuracy” remains a challenge, with more than two-thirds of those surveyed citing what amounts to “extracting signal from noise.” Respondents also listed, in descending order, root cause analysis, eliminating redundant tasks and “understanding application to infrastructure dependencies.”
The last item is bound to grow in urgency as enterprises shift to micro-services such as lightweight application containers used to efficiently deploy distributed apps.
IT teams are turning to AIOps “to help them sift through huge volumes of IT infrastructure data,” Mahesh Ramachandran, OpsRamp’s vice president of product management, noted in a contributed article in early April.
“From events and logs to telemetry data or other workload performance and availability information, IT leaders are seeking comprehensive solutions for automating many of the most time- and resource-intensive ITOps activities to free their teams to take a more strategic role in driving digital innovation and transformation,” Ramachandran added.
The OpsRamp survey concludes that the leading use cases driving AIOps adoption are intelligent alert notifications, root cause analysis, and anomaly detection. “These three use cases are a great test bed for demonstrating the effectiveness of AIOps tools and will let IT leaders make the business case for gradually expanding the scope,” the survey added.