Cloudian enhances AI processing with NVIDIA GPUDirect tie-up

Cloudian has announced its integration with NVIDIA's GPUDirect storage technology, which delivers performance speeds exceeding 200GB/s.

The collaboration between Cloudian and NVIDIA aims to simplify AI training and inference management at a large scale while reducing costs typically associated with complex data migrations. The new integration is reported to significantly decrease CPU overhead by 45% during data transfers, freeing up computational resources that can be redirected to AI processing tasks.

David Small, Group Technology Officer at Softsource vBridge, shared his perspective on the integration: "We've seen firsthand how data management challenges can hinder AI adoption. Cloudian's GPUDirect for Object Storage will simplify the entire AI data lifecycle, which could be the key to democratising AI across various business sectors, allowing companies of all sizes to harness the power of their data. We're particularly excited about how this could accelerate AI projects for our mid-market clients who have previously found enterprise AI solutions out of reach."

Michael Tso, Chief Executive Officer of Cloudian, expressed the company's commitment to transforming AI data workflows: "Cloudian is proud to be at the forefront of transforming how enterprises and AI hyperscalers harness data to realise the power of AI. For too long, AI users have been saddled with the unnecessary complexity and performance bottlenecks of legacy storage solutions. With GPUDirect Storage integration, we are enabling AI workflows to directly leverage a simply scalable storage architecture so organisations can unleash the full potential of their data."

From a technological standpoint, Cloudian's HyperStore system now incorporates features such as limitless scalability to accommodate the growing demands of AI datasets. It eliminates the need for complex data migrations by directly supporting AI workflows on the existing data. The architecture also ensures consistent and scalable performance, tested to over 200GB/s as measured on the industry-standard GOSBench benchmark.

Michael McNerney, Senior Vice President of Marketing and Network Security at Supermicro, acknowledged the advancements achieved in collaboration with Cloudian: "At Supermicro, we're committed to delivering the most advanced and efficient solutions for AI and deep learning. Cloudian's integration of NVIDIA GPUDirect Storage with the HyperStore line of object storage appliances based on Supermicro systems represents a significant innovation in the use of object storage for AI workloads. This will enable our mutual customers to deploy more powerful and cost-effective AI infrastructure at scale."

Rob Davis, Vice President of Storage Technology at NVIDIA, highlighted the role of performance in AI applications: "Fast, consistent, and scalable performance in object storage systems is crucial for AI workflows. It enables real-time processing and decision-making, which are essential for applications like fraud detection and personalised recommendations."

The integration further minimises the time and expense of managing large AI datasets by eliminating separate file storage layers. This reduction not only simplifies the data management process but also provides a single, unified data lake that reduces operational costs. Moreover, the deployment involves no vendor-driven kernel modifications, mitigating potential vulnerabilities often associated with kernel changes.

The Cloudian HyperStore's architecture incorporates integrated metadata, which enables quick data searches and retrievals without relying on external databases, expediting AI training and inference processes.

With Cloudian's comprehensive security features, enterprises employing AI can manage sensitive data with various tools. These include access controls, encryption protocols, integrated key management, and S3 Object Lock for protection against ransomware, ensuring data security throughout its lifecycle.

Cloudian's efforts in integrating with NVIDIA's GPUDirect Storage highlight a strategic step towards harnessing GPU capabilities for efficient AI processing. This provides a scalable and secure platform for enterprises to capitalise on the potential of AI data.