AIRI//S is a scalable AI infrastructure that integrates Pure’s scale-out storage with Nvidia compute systems. Credit: Zapp2Photo / Getty Images Pure Storage announced updates to its AIRI//S line of AI-ready infrastructure, which it co-developed with Nvidia. The two vendors launched AIRI in 2018, claiming it was the first AI-oriented reference architecture that simplified the process of building an AI infrastructure by connecting compute with storage. AIRI is essentially a combination of Pure’s scale-out FlashBlade//S and Nvidia’s DGX ultra-dense GPU box. Pure provided the storage, and Nvidia provided the compute. This latest move, unveiled at the Pure//Accelerate techfest22 conference in Los Angeles, is quite an advancement, however. The new release of AIRI//S is powered by Nvidia DGX A100 systems, featuring end-to-end networking provided by Nvidia’s Quantum InfiniBand and Spectrum networking. A DGX A100 system comes with eight Ampere-generation A100 GPUs and up to ten ConnectX-6 network adapters from Mellanox. Pure Storage Support for Nvidia’s Magnum IO GPUDirect Storage is planned in future updates. GPUDirect allows SSD storage to communicate directly with the memory on the GPU rather than having to go through the CPU and main memory, thus lessening the burden on the CPU. “Traditional approaches to AI infrastructure often result in silos of servers and storage that are either over-spent on capacity or starve AI workloads. With a focus on simplicity and scalability, AIRI//S enables global enterprises to achieve better time to insights and make the most out of their data with AI,” said Amy Fowler, vice president, strategy and solutions, FlashBlade at Pure Storage, in a statement. FlashBlade//S gets an update The AIRI news isn’t the only news out of Pure//Accelerate techfest22. The FlashBlade//S product family of file and object storage arrays has been updated with a new modular architecture built on co-designed hardware and software. The new platform offers more than double the density, performance and power efficiency of previous versions thanks to its modular architecture, which decouples compute, storage, and networking elements. Because the FlashBlade//S disaggregates the elements, customers can scale compute and capacity separately as needed. Both can be upgraded independently and non-disruptively, so you can add more compute to one array if needed and more capacity to another. FlashBlade//S is sold separately and as part of AIRI//S. Consumption-based service updates Lastly, Pure has streamlined its Evergreen subscription service offerings. It launched Evergreen in 2015 as a software subscription contract that includes hardware and software upgrades. It set the stage for the Pure-as-a-Service consumption model launched in 2018. There are now three Evergreen offerings: Evergreen//Forever (formerly Evergreen Gold), the traditional model where the customer owns the appliance and has a subscription to software and hardware in-place updates at regular intervals; Evergreen//One (formerly Pure-as-a-Service), the consumption-based service model for storage; and Evergreen//Flex, where Pure owns the whole batch of arrays in an on-premises managed services model. Related content news High-bandwidth memory nearly sold out until 2026 While it might be tempting to blame Nvidia for the shortage of HBM, it’s not alone in driving high-performance computing and demand for the memory HPC requires. By Andy Patrizio May 13, 2024 3 mins CPUs and Processors High-Performance Computing Data Center news CHIPS Act to fund $285 million for semiconductor digital twins Plans call for building an institute to develop digital twins for semiconductor manufacturing and share resources among chip developers. By Andy Patrizio May 10, 2024 3 mins CPUs and Processors Data Center news HPE launches storage system for HPC and AI clusters The HPE Cray Storage Systems C500 is tuned to avoid I/O bottlenecks and offers a lower entry price than Cray systems designed for top supercomputers. By Andy Patrizio May 07, 2024 3 mins Supercomputers Enterprise Storage Data Center news Lenovo ships all-AMD AI systems New systems are designed to support generative AI and on-prem Azure. By Andy Patrizio Apr 30, 2024 3 mins CPUs and Processors Data Center PODCASTS VIDEOS RESOURCES EVENTS NEWSLETTERS Newsletter Promo Module Test Description for newsletter promo module. Please enter a valid email address Subscribe