Cisco launches Data Fabric to unify machine data for AI insights
The Cisco Data Fabric has been announced, offering an architecture for organisations to process machine data and prepare it for use in artificial intelligence applications.
Cisco states the Data Fabric will enable companies to reduce the challenges associated with managing large volumes of machine data and make it possible to use that data for AI model training, automated agent workflows, and operations analysis.
"Organisations everywhere are sitting on a gold mine of machine data that's been too complex, cumbersome, and costly to leverage for AI, until today. From sensor readings and factory metrics to checkout system data and event updates from apps, servers, networks and more, machine data drives how businesses operate. Splunk revolutionized data and analytics for the cloud. And now, the Cisco Data Fabric is poised to do the same for AI by making it possible for enterprises to build AI models with their own proprietary machine data," said Jeetu Patel, President and Chief Product Officer at Cisco.
The Data Fabric is based on the Splunk platform, which Cisco acquired. Its stated function is to unify machine data from across enterprise IT and operational environments, including data generated from networks, servers, applications, and physical sensors. Cisco reports that the platform is built to address the high cost and complexity organisations encounter when attempting to centralise and analyse distributed machine data for AI-driven decision-making.
"Our goal is to give customers the fastest, most secure path from data to action. By embedding AI across the platform and embracing open standards, we're not just helping organizations analyze information faster - we're enabling them to anticipate change, scale innovation without unnecessary complexity, and deliver digital services that are more resilient, adaptive, and responsive to the needs of their users," said Kamal Hathi, SVP and GM, Splunk, a Cisco company.
Data intelligence
Cisco outlines the Data Fabric's capabilities include advanced data filtering, shaping, and tiering, as well as federated analytics that can bring together insights from multiple data sources. According to Cisco, intelligent edge data management creates near real-time operational intelligence, while an AI-driven experience layer aims to reduce manual overhead and speed up decision-making by teams.
Among the features highlighted are the ability to operate on machine data at what Cisco terms "extreme scale", supporting transformation across edge, cloud, and on-premises environments. This includes real-time searching and analysis of data residing on cloud data lakes as well as traditional storage platforms, with support for Amazon S3 now, and Apache Iceberg, Delta Lake (with Spark), Snowflake, and Microsoft Azure planned for the future. Additional data integrations are expected to follow in 2026.
The platform is described as "flexible" and compatible with open industry standards, allowing for deployments in both private and public cloud environments and supporting open, self-service integrations to support innovation.
Proprietary data for AI
Cisco notes that the Data Fabric includes a Time Series Foundation Model (to be released in November 2025), bringing pattern analysis and temporal reasoning capabilities for detection of anomalies and analysis of operational incidents. The Splunk Machine Data Lake and associated AI tools are available immediately, and provide the environment for both model training and analytics on machine data.
Additional resources such as the Splunk AI Toolkit and Model Context Protocol Server are positioned to help convert machine data into what the company calls a "fuel source for advanced AI capabilities." Cisco expects these tools to play a role in automated root cause analysis, forecasting, and proactive incident response.
User and agent experience
To support teams and AI agents, Cisco is introducing the Cisco AI Canvas, which integrates with the Splunk Cloud Platform. This feature will create virtual spaces (described as virtual war rooms) for collaborative analysis, access to investigation and visualisation tools, and streamlined sharing of reports and findings within a unified environment. Cisco reports that the platform is AI-native, aiming to provide support at every stage of the data lifecycle, from ingest to cooperative analysis and reporting.
"The Cisco Data Fabric addresses a critical pain point in today's AI-driven enterprise: the challenge to quickly and securely unify vast streams of machine data for continuous resilience. By enabling a federated approach that eliminates data movement, it provides a pragmatic solution for organizations operationalizing AI at scale. Its focus on real-time search, coupled with a repository for AI-ready data, provides tangible value by reducing complexity and time to insights. This unified architecture is a strong step toward helping customers build more resilient and trustworthy AI systems," said Archana Venkatraman, Senior Research Director, Cloud Data Management, IDC.
Product availability
Cisco states that the Data Fabric is based on Splunk Enterprise and Splunk Cloud Platform, with capabilities available currently and additional features planned into 2026. The AI Toolkit is available now, with hosted models expected in 2026. Federated analytics using Replay S3 will be enabled in October 2025, the Time Series Foundation Model will be published on Hugging Face in November 2025, and the full Cisco AI Canvas integration will become available in 2026.
