Graid Technology Inc. logo in front of a shiny blue decorative background.
Video
October 4, 2021

VIDEO: StorageReview.com - Graid Technology NVMe RAID Card Review (It's really AI and a GPU)

"Mind blowing numbers...  Graid SupremeRAID™ performed exceedingly well; certainly worth a POC to see what this card can do for your high performance applications."

"We recently posted a written review on the site of the  Graid SupremeRAID SR-1000. Ostensibly an NVMe RAID card, that notion gets dispelled pretty quickly upon unboxing. It's actually an NVIDIA GPU that pairs with GRAID's AI to deliver lossless RAID for NVMe SSDs. This is a huge deal because right now, the other options aren't great. Software RAID has many limitations, as do the popular tri-mode RAID cards that dominate today. Join us to learn more about the  Graid solution in this video review." Read the full review here.

Learn More

News & Resources

Join Graid Technology at the 2026 Dell Technologies Forum Shanghai on August 21! Discover how Dell × SupremeRAID™ solutions power AI from cloud to edge—from private AI and HPC clusters to edge AI and KV Cache offloading—with 2–10× higher storage performance, up to 80 GB/s per node, and up to 24× faster TTFT.
Don't miss our joint speaker session with Supermicro, where we'll walk through scaling approaches across every tier, from single-server to rack-scale to monolithic deployments, and show how dense NVMe GPU platforms paired with SupremeRAID™ turn SSDs into a high-performance, protected KV cache tier. If storage is becoming your AI infrastructure bottleneck, this is the session to catch. See you at FMS 2026!
Graid Technology has achieved an industry-first benchmark: 100 million IOPS on a protected RAID 5 volume for GPU-initiated I/O, built from 32 KIOXIA XD8 NVMe SSDs. The result shows that resilient NVMe storage can now operate at the scale that current and future AI deployments require, which is especially significant for environments that require both extreme performance and enterprise-class data protection. In benchmark testing with a WholeGraph training workload, [...]