Get the SupremeRAID™ Summer 6-Pack! Buy 5 licenses, get the 6th FREE / Read the Storage Tax Blog
Contact Us
Solution Brief
September 23, 2025
Unlock Extreme NVMe Performance: Award-winning SupremeRAID™ HE with Supermicro & BeeGFS
In today’s data-intensive world — from AI training to scientific computing to large-scale analytics — speed, reliability, and cost efficiency are non-negotiable. That’s where SupremeRAID™ HE shines. Developed by Graid Technology in partnership with Supermicro and BeeGFS, this high-performance NVMe RAID solution delivers the kind of throughput and scalability that today’s storage infrastructures demand.
What Makes It Different
- GPU-Accelerated RAID: SupremeRAID™ HE offloads overhead from the CPU by using GPU parallelism, letting your compute resources focus on what matters most.
- Cross-Node High Availability & Array Migration: Rather than replicating data across nodes (which wastes resources), this solution enables seamless array migration across nodes for redundancy and resilience.
- Near-Theoretical NVMe Throughput: Tested on a 2U Supermicro SBB system (SSG-221E-DN2R24R) with 24 dual-port NVMe drives, the brief shows read speeds of 132 GB/s and write speeds of 83 GB/s (post-RAID) locally — that’s over 90% of theoretical max.
- Network Saturation on 400 Gb/s Links: On the client side via BeeGFS, the system pushes nearly 93 GB/s reads and 84 GB/s writes, saturating 400 Gigabit networking.
Why It Matters
For organizations running HPC clusters, intensive analytics, or any workload where storage truly is the bottleneck, SupremeRAID™ HE represents a leap forward. You get:
- Better performance without dramatic increases in cost
- Higher availability without the overhead of full cross-node replication
- Scalability in a compact 2U footprint
Ready to see the numbers & technical details? Download the full solution brief now to explore testing environments, configuration options, and performance benchmarks. And if you’re thinking about deploying this for your environment, contact us — our team at Graid Technology is ready to help with sales inquiries, custom configurations, and integration guidance.
News & Resources
September 14, 2026
Breaking the KV Cache Inference Wall — Solidigm Data Insights Podcast, Featuring Graid Technology
"You've got a four-lane highway, and to get to your storage, you have to go down to a one-lane bridge." Garrett McKibben joins Solidigm on TechArena's Data Insights podcast to unpack why KV cache overflow is choking AI inference.
.jpg)
Graid Technology CEO Leander Yu and VP of Product Development Alven Yen explain how moving RAID processing onto the GPU breaks the bottleneck traditional controllers cannot solve, delivering 36 million GPU-initiated IOPS for AI infrastructure.
.jpg)
Join Graid Technology at Dell Technologies Forum Taipei 2026 and discover how SupremeRAID™ AE transforms local NVMe storage into a high-performance, scalable, and protected extended KV cache—reducing redundant computation and accelerating time to first token (TTFT).