Pure Storage has announced new capabilities in its platform to improve users’ ability to deploy AI, improve cyber resilience and modernise their applications.
Headlining the developments were what the company described as first of its kind storage automation, dubbed the next generation of Pure Fusion, to unify arrays and optimise storage pools on the fly across structured and unstructured data, on-premises and in the cloud.
This will be available across the company’s entire platform to all global customers, including managed service providers.
A generative AI copilot for storage is also being delivered, providing a new way to manage and protect data using natural language.
Pure Storage said this new AI capability builds on data insights from tens of thousands of customers to guide storage teams through the investigation of complex performance and management issues and to stay ahead of security incidents.
Pure Storage touts its platform as-a-service model as delivering agility and risk reduction to organisations with a simple, consistent storage architecture that addresses a broad set of use cases across on-premises, public cloud, and hosted environments.
It also promises continuous and non-disruptive upgrades married to service level agreements so reliability, performance and sustainability requirements can be met and demonstrated.
A new Evergreen One site rebalance SLA could also prove popular among MSPs as well as end users.
If capacity is no longer needed at a specific site, if a data centre is being consolidated or if performance at one site was oversized, that can now be used elsewhere through a rebalance option once every 12 months for each subscription.
Pure Storage is introducing other innovations in the platform to help businesses accelerate successful AI deployments and to future proof their environments.
First, a new Evergreen One for AI storage-as-a-service option with a new service level agreement that provides guaranteed storage performance for GPUs to support training, inference and high performance computing workloads introducing the ability to purchase based on dynamic performance and throughput needs.
Mark Jobbins, chief technology officer Pure Storage Asia-Pacific and Japan, told Reseller News Evergreen One was traditionally split into two – block and file or unified fast file and object – and then tiered based on a performance profile.
“What we’ve observed in the world of AI is it flips it around a different way,” Jobbins said.
“This is typically in object format but people want throughput. The new service agreement Evergreen One for AI allows customers to subscribe to a throughput service level agreement to drive the GPU-based compute layer.”
It was envisaged MSPs could also subscribe to this and pass that on to their end users.
A second innovation, secure application workspaces with fine-grained access controls, makes storage infrastructure transparent to application owners, who gain fully automated access to AI innovation without sacrificing security, independence, and control.
These combine Kubernetes container management, secure multi-tenancy and policy governance tools to enable advanced data integrations between enterprise mission-critical data and AI clusters.
Pure Storage is also expanding and accelerating AI adoption with Nvidea DGX SuperPOD certification expected by the end of the year.
“Enterprises adopting AI need storage performance and flexibility as they architect their infrastructure to address AI workloads at scale,” said Charlie Boyle, vice president of DGX Platform, at Nvidea.
“Pure Storage’s certification with the NVIDIA DGX and OVX platforms helps deliver highly-performant solutions for customers at every stage of their AI journey.”
Also announced at the vendor’s Pure//Accelerate conference this month were new cyber resilience capabilities including SafeMode for ransomware recovery and an enhanced cyber recovery and resilience service level agreement that now also covers disaster recovery.
