GreenLake Intelligence helps IT groups handle operations (Picture courtesy HPE)
HPE right this moment launched GreenLake Intelligence, a brand new AI-powered framework designed to proactively monitor IT operations and autonomously take motion to stop issues earlier than they come up. The corporate, which made the announcement at its annual HPE Uncover convention in Las Vegas, says the shift to agentic AIOps will alleviate the burden on human directors and operators whereas nonetheless maintaining them within the loop.
Organizations right this moment are struggling to watch and management the huge array of servers, storage, networks, working programs, databases, and functions that they depend on to ship digital companies to their prospects. Conventional monitoring instruments have lengthy since been overwhelmed by the tsunami of observability knowledge flowing off these programs, which led on to the reliance on machine studying strategies to detect abnormalities, which led to the AIOps period.
However the quantity of observability knowledge has continued to develop unabated, and now the AIOps and observability software distributors have gotten overwhelmed. In some conditions, the scale of firms’ observability knowledge rivals the enterprise knowledge that really drives their enterprise. To handle this example, some firms are turning to agentic AI to not solely monitor IT operations however to imitate human responses to them.
That is the trail that HPE is taking with GreenLake Intelligence, which is certainly one of a number of bulletins HPE made at its Uncover convention right this moment. GreenLake Intelligence is a brand new providing underneath HPE’s GreenLake Copilot that routinely deploys AI brokers to watch an array of server, storage, networking, and virtualized assets, together with buyer IT assets developed by HPE in addition to third-party gear, operating both on prem or within the cloud.
GreenLake Intelligence is predicated on a domain-specific massive language mannequin (LLM) that’s educated on HPE’s telemetry, logs, configurations, and help knowledge. It additionally consists of an array of varied brokers that use Mannequin Context Protocol (MCP) to coordinate with one another by means of an “agentic mesh.”
GreenLake Intelligence represents HPE’s first foray into agentic AIOps (Picture courtesy HPE)
“We’re constructing Greenlake brokers with the situational consciousness they should make clever choices primarily based on real-time infrastructure metrics and enterprise logic,” writes Brian Gruttadauria, the CTO of HPE’s hybrid cloud/AI enterprise in a weblog submit. “Whether or not you’re making an attempt to retrieve knowledge from an HPE Alletra Storage MP X10000 storage array or provision a Kubernetes cluster for a mission-critical software, or perceive the expertise of a person on Wi-Fi, the brokers orchestrate to ship on the duty requested.”
The usage of MCP permits the brokers to speak with one another, calling in further brokers to resolve AIOps challenges when the necessity arises. As an illustration, if an agent detects anomalous habits in monetary knowledge, it may request enhanced assets from one other agent accountable for dynamic workload orchestration, HPE says.
GreenLake Intelligence will assist overworked IT groups by shouldering a few of the operational burden, HPE says. The copilot will assist groups to maintain issues straight in more and more advanced hybrid cloud deployments, and likewise help with root trigger intelligence, the corporate says. In brief, it’s going to function the connective tissue between human operators and the programs they management by enabling natural-language interactions, offering contextual insights, and offering suggestions.
“HPE is reimagining hybrid IT as solely we will do, catapulting organizations from the period of hybrid complexity to the period of agentic-AI-powered cloud operations,” acknowledged HPE President and CEO Antonio Neri. “HPE’s new imaginative and prescient for hybrid IT is fueled by agentic intelligence at each layer of infrastructure, so enterprises can notice their boldest ambitions and obtain beforehand unattainable ranges of IT operations efficiency and effectivity.”
Associated Objects:
Decoding Metrics: Why Observability’s ‘Common Language’ Nonetheless Wants Translation
2025 Observability Predictions and Observations
Explosion of Observability Knowledge from Cloud Reaches Tipping Level, Dynatrace Says
