INTELLIGENT SOFTWARE

AST Data , the AI data platform company , unveiled a new AI cloud architecture designed to deliver unprecedented levels of performance , quality of service , zero-trust security and space , cost , power efficiency for the AI factory . Building on NVIDIA BlueField-3 data processing unit , DPU technology , VAST Data ’ s parallel system architecture makes it possible to disaggregate the entirety of VAST ’ s operating system natively into AI computing machinery , transforming supercomputers into AI data engines .

The NVIDIA BlueField networking platform combines robust compute power and integrated hardware accelerators to create secure and software-defined accelerated computing infrastructure for AI . By outfitting each GPU server with a dedicated NVIDIA BlueField DPU running a stateless container

Jeff Denworth , Co-founder VAST Data that powers the VAST parallel services operating system , this new architecture design embeds storage and database processing services directly into AI servers and delivers true linear data services designed to scale to hundreds of thousands of GPUs .

This new VAST architecture , running VAST software on BlueField DPUs in the AI servers , is being evaluated and deployed first at CoreWeave , the specialised GPU cloud provider . VAST and CoreWeave began partnering in 2023 to build some of the world ’ s most scalable AI machinery and to help many of the world ’ s LLM builders and blue-chip enterprise customers build their own AI factories .

Moreover , by removing multiple layers of x86 hardware and networking from VAST ’ s network-attached Data Platform infrastructure , this new AI factory architecture dramatically reduces the cost , footprint , and power associated with AI data services .

“ We are extremely proud to partner with NVIDIA to help industrialise AI computing ,” said Jeff Denworth , co-founder at VAST Data . “ This new architecture is the showcase to express the parallelism of the VAST Data Platform .”

VAST ’ s Disaggregated , Shared Everything , DASE architecture leverages the processing power of NVIDIA BlueField-3 to require less independent compute and networking resources , reducing the power usage and data centre footprint for VAST infrastructure by 70 %.

The combined end-to-end solution results in a net energy consumption savings of over 5 % compared to deploying NVIDIA-powered supercomputers with the previous VAST distributed data services infrastructure .

By providing each GPU server with a dedicated and truly parallel storage and database container , this new AI factory architecture eliminates contention for data services infrastructure . VAST ’ s DASE architecture features extreme parallelism such that each NVIDIA BlueField-3 can read and write into shared namespaces of the VAST Data Platform without coordinating IO across containers .

In essence , this architecture eliminates infrastructure contention at the most fundamental level . This contention-less architecture is essential for multi-tenant service providers who need to meet the contractual Service Level Objectives of their clients while also maximising the utilisation of all GPU computing assets .

This new AI factory architecture ensures that data and data management remain protected and isolated from host operating systems . Compared to AI computers that use parallel file system clients , which have an intimate understanding of the data services layer , VAST is able to eliminate many attack vectors . •

INTELLIGENT TECH CHANNELS 4949