Lenovo OEMing latest NetApp EF-Series block arrays as NetApp's Novus hits the ground running

Lenovo has launched the ThinkSystem DE6500 and DE6800 block storage arrays, OEM versions of NetApp's EF50 and EF80 all-flash systems announced in March, which deliver over 110 GBps read and 55 GBps write bandwidth with up to 1.5 PB of capacity in a 2 RU chassis. The DE6800 offers roughly 3× the IOPS and ~2.7× the throughput of the DE6500, with higher-speed host ports including 200Gb, and both run Lenovo's SAN OS and System Manager. NetApp also said its disaggregated ONTAP storage variant Novus is now generally available, architected for up to 100 TB per second of aggregate throughput, and is being deployed in Nebul's Netherlands-based European sovereign cloud.

AI/ML Lenovo OEMing latest NetApp EF-Series block arrays as NetApp's Novus hits the ground running Lenovo has announced its DE6500 and DE6800 block storage arrays, based on NetApp’s EF50 and EF80 systems, while NetApp is promoting its Novus super-fast ONTAP system for zettabyte-scale AI factories. It announced these arrays alongside high-performance, Neptune liquid-cooled servers. The Lenovo ThinkSystem DE6800 and DE6500 arrays are designedfor scale and performance, and optimized for high-throughput, latency-sensitive block workloads. The DE6800 is well suited for high throughput applications demanding reliable storage, while the DE6500 delivers economical high performance with simplified deployment and management. Dan McNamara, SVP and GM, Compute and Enterprise AI, AMD, said: “As AI and HPC workloads converge, organizations need infrastructure that delivers exceptional compute performance, energy efficiency and flexibility at scale. The new Lenovo ThinkSystem platforms powered by 6th Gen AMD EPYC Server CPUs help customers accelerate demanding workloads – from engineering and scientific research to AI inferencing and agentic AI – while building a more efficient, scalable foundation for continued innovation.” NetApp’s EF-series systems are all-flash arrays with 24 NVMe drive bays, running the Santricity OS, which provide fast block storage without the added features and complexity of NetApp’s ONTAP-based, all-flash SAN ASA , block-access-only A-Series. The EF50 and EF80 https://www.blocksandfiles.com/ai-ml/2026/03/17/nvidia-gtc-storage-news-roundup/5209446 were announced in March and provide over 110 GBps read and 55 GBps write bandwidth, with up to 1.5 PB of capacity, in a 2 RU chassis. In DE6500 and DE6800 guise they run the Lenovo SAN OS and System Manager. They are described like this: - DE6500: Entry-to-midrange platform for mixed workloads. Well-suited to databases and analytics, backup/recovery, general SMB, low-latency apps, video surveillance, and technical computing. - DE6800: Higher-performance system with roughly 3× the IOPS and ~2.7× the throughput of the DE6500, plus more memory and higher-speed host ports including 200Gb . Targeted at AI, big data analytics, Oracle/SQL databases, HPC, and real-time analytics Get detailed information from Lenovo DE6500 https://lenovopress.lenovo.com/lp2436-lenovo-thinksystem-de6500-storage-array and DE6800 https://lenovopress.lenovo.com/lp2437-lenovo-thinksystem-de6800-storage-array product guides. Novus NetApp is now discussing the deployment of its Novus https://www.blocksandfiles.com/file/2026/09/29/netapp-unveils-fastest-file-system-on-the-planet-with-great-raft-of-other-announcements/5299808 disaggregated ONTAP storage variant in the Netherlands-based Nebul https://www.blocksandfiles.com/object/2026/10/26/neocloud-nebuls-five-storage-options/5305288 European sovereign cloud. Nebul also uses DDN EXAScaler and Infinia, NetApp ONTAP and VAST Data’s AI OS. Novus is now generally available, designed to maximize utilization for millions of GPUs, and is architected for up to 100 TB per second of aggregate throughput. It is a tad late to the high-performance, AI factory, storage software market, and has to compete against established parallel file system players such as DDN’s EXAScaler Lustre system, IBM’s Spectrum Scale, Quobyte, VDURA and BeeGFS, Dell’s Lightning and Hammerspace’s GDE, newer, disaggregated HW and object-style DDN Infinia, HPE Alletra MP, Everpure EXA, WEKA’s Neural Mesh, and VAST Data’s AI OS. It’s significant advantage is that it is ONTAP-compatible and thus should be easier for NetApp shops to adopt than any of the other systems mentioned. Willem Hendrickx, SVP and GM, EMEA and LATAM at NetApp, said: “Neoclouds and GPU-as-a-Service providers are building AI factories at a scale that traditional storage architectures were never designed to support. As demand for data sovereignty grows, AI providers like Nebul must also build their infrastructure to address the compliance, security and multi-tenancy requirements of regulated industries such as government, healthcare and biotech. “Novus combines a cutting-edge architecture built from the ground up for modern AI factories with the NetApp Platform’s established capabilities for making data AI-ready and secure, supporting customers with a faster, lower-risk path from AI pilot to production.” NetApp says Novus is already gaining momentum with neocloud providers to explore how NetApp and AI cloud providers can bring NetApp-powered AI infrastructure to customers around the world. Find out more here https://www.netapp.com/novus/ . It will be interesting to see if Lenovo adopts Novus and adds it to its AI storage portfolio.