Cohesity expands Gaia for on-premises backup AI search

Cohesity has expanded its Gaia enterprise knowledge discovery assistant to include AI-powered search capabilities for on-premises backup data.

The new features have been developed in collaboration with NVIDIA using its accelerated computing and AI Enterprise software platform, including NVIDIA NIM microservices and the NVIDIA NeMo Retriever. The Cohesity Gaia solution is now validated for use with Cisco UCS, Hewlett Packard Enterprise (HPE), and Nutanix hardware, giving organisations a choice of supported infrastructure platforms.

Many enterprises now favour hybrid cloud solutions but prefer to retain some data on-premises due to requirements around security, compliance, and performance. The updated Gaia allows them to leverage generative AI tools while keeping sensitive backup data within their infrastructure, without transferring it to a cloud environment.

"Many of our customers require an AI solution for on-premises environments. This solution allows them to maintain complete control over sensitive data while benefiting from generative AI capabilities at the same time," said Sanjay Poonen, President and Chief Executive Officer, Cohesity. "By deploying Cohesity Gaia on-premises, they can harness powerful data intelligence directly within their environment and not worry about any of that data leaving their infrastructure."

The integration of NVIDIA's AI technologies with Cohesity Gaia brings generative AI capabilities to backup and archived data, allowing users to search, analyse, and extract insights from large volumes of stored information. This development makes it possible for enterprises to index and query data in multiple languages, tailor environments to specific needs, and perform searches across petabyte-sized datasets.

Pat Lee, Vice President, Strategic Enterprise Partnerships, NVIDIA, commented: "Access to data is a crucial foundation for building advanced AI systems, including agentic AI. By integrating NVIDIA NIM microservices and NeMo Retriever into Cohesity Gaia, enterprises can harness AI-driven insights directly within their infrastructure to preserve data accessibility and security while unlocking new levels of intelligence."

Paul Henaghan, Managing Director for Australia and New Zealand, Cohesity, said: "Australian companies recognise that their data – primary and secondary – enabled through AI creates significant competitive advantage. Cohesity Gaia's analytics capabilities drive data-driven decisions and innovation. By extending Cohesity Gaia to data stored on-premises, it ensures compliance with Australian data sovereignty laws and industry regulations, delivering security and control, while maximising the value of secondary data."

The Cohesity Gaia solution for on-premises environments offers a number of practical advantages, according to the company. These benefits include maintaining control and security of data, using high-performance AI underpinned by NVIDIA technology, scalable and customisable infrastructure, and reference architectures for simplified deployment. The solution includes pre-packaged on-premises large language models (LLMs), so no customer backup data needs to be moved to the cloud for analysis.

One customer currently assessing the potential of Cohesity Gaia's on-premises solution is JSR Corporation, an international research and manufacturing business. "With research centres in many locations, our researchers speak different languages. We want to be able to use generative AI to look across years of our data to discover insights that can accelerate the rate of our research and discovery," said Ryan Reed, Head of IT, JSR Corporation. "Cohesity is leveraging NVIDIA AI Enterprise software and expertise to make that possible with an on-premises AI solution that uses our own collective knowledge."

Cohesity is working with Cisco, HPE, and Nutanix to deliver Gaia to enterprises through validated integration with each vendor's solution stack. Cisco and Cohesity will offer Gaia software on Cisco AI PODs, while HPE will include deployment and validation with its HPE Private Cloud AI platform. Nutanix support will be delivered through integration with its enterprise AI infrastructure.

Jeremy Foster, SVP and GM of Cisco Compute, stated: "Cohesity Gaia's on-premises solution combined with Cisco's AI-optimised infrastructure would enable organisations to unlock the insights and untapped value in their backup data. It promises to also deliver the performance and scalability needed for growing AI workloads, while keeping sensitive data within an organisation's control."

Patrick Osborne, SVP of Technology Acceleration, HPE, explained: "To accelerate value creation with AI, enterprises need exceptional performance and simplicity across their data estate. Cohesity Gaia paired with HPE AI-optimised infrastructure, such as HPE Private Cloud AI, provides both. These integrated solutions featuring HPE storage and servers let organisations efficiently manage unstructured data growth, reduce costs and complexity, and create new insights within their own secure environments."

Thomas Cornely, SVP of Product Management, Nutanix, said: "Cohesity Gaia integrated with Nutanix Enterprise AI infrastructure will enable enterprise customers to efficiently derive value from backup data. This joint solution will deliver valuable AI powered insights with privacy, control, and efficiency, all within the confines of their on-premises environments."

Cohesity Gaia for on-premises environments is expected to be available for general adoption in mid-2025.