NetApp has rolled out its next-generation EF-Series storage lineup, the EF50 and EF80, aimed at high-performance workloads such as AI, high-performance computing (HPC) and transactional databases.
The new systems are built to handle growing demands from enterprises and neocloud providers, including use cases such as sovereign AI clouds and AI-driven manufacturing.
NetApp says the EF50 and EF80 deliver over 110GBps read throughput and 55GBps write throughput, a 250% improvement over previous generations. They offer 63.7GBps per kW in power efficiency and up to 1.5PB of storage in a 2U chassis, enabling high-throughput low-latency workloads without occupying excessive data centre space.
Senior Vice President and General Manager of Enterprise Storage Sandeep Singh said:
“As businesses contend with ever-increasing data volumes and performance-intensive applications such as AI model training, AI inferencing and high-performance computing, they need infrastructure that delivers speed, scalability and efficiency without added complexity.”
The EF-Series can work with parallel file systems such as Lustre and BeeGFS, allowing HPC simulations to run faster and keeping GPUs fully utilised, which is critical for AI training and inference.
NetApp emphasised that the EF-Series supports the entire AI data pipeline from ingesting and preparing data to feeding generative AI models that produce business insights.
Industry experts highlighted the performance benefits of the new EF-Series:
- Clayton Vipond, Senior Solution Architect at CDW said:
“The refreshed NetApp EF-Series delivers the throughput and capacity businesses need to scale high-powered workloads that transform data into insights and outcomes.” - Sumeet Arora, Chief Product Officer at Teradata said:
“The improved models may help reduce infrastructure complexity while supporting AI and data modernisation initiatives.” - Simon Robinson, Principal Analyst at Omdia said:
“With the new EF-Series systems, NetApp is delivering a solution that addresses the performance needs of large-scale AI projects, whether model training or inference.”
The EF-Series builds on decades of proven reliability, with over one million installations worldwide, now optimised for AI, HPC and other data-intensive workloads that demand speed and efficiency.
