Overcome Performance Bottlenecks with Software-Defined RAID


Enterprises are facing performance issues with SSDs, but GRAID Technology is working to change that.

Data Startup GRAID Technology Offers Software-Defined Solution That Helps Enterprises Take Advantage of Large Scale RAID Technology

Software-defined storage is proving to be in high demand in the tech industry, with SSDs bringing instantly accessible storage to enterprises gradually replacing HDDs for increased scalability and flexibility. Gartner predicts this trend will continue, with enterprise SSDs accounting for nearly one-third (32%) of total HDD/SSD space by 2026.

SSDs are often implemented with RAID (Redundant Array of Independent Disks) – a data storage virtualization technology that combines multiple physical disk components into one or more logical units, providing cost savings and increased performance. The resilience of this technology is known to play an important role in data protection.

However, the use of this storage technology frequently faces challenges that create bottlenecks and stifle innovation. The evolution of flash storage performance leaves existing storage architectures behind, meaning businesses can’t take advantage of it, while software RAID competes for computing power with other applications, slowing operations.

Additionally, hardware RAID is often limited to less than 10 SSDs, which does not deliver the full performance potential of the overall application.

Source: GRAID Technology

Eradicate the bottleneck

To address this challenge, California-based data startup GRAID Technology, which was speaking at the latest IT Press Tour in Silicon Valley, offers a software-defined solution named SupremeRAID, which can facilitate the provision of computing power for up to 32 SSDs.

With a composable design, the solution can be implemented out of the box and sits on the GPU. The off-path protection provided by the solution allows data to travel directly from CPU to storage without passing through the SupremeRAID card. Its AI component, meanwhile, can process 19 million input/output operations per second (IOPS) and 7 GB per second. All of this helps solve the flash storage bottleneck that businesses often face.

Source: GRAID Technology

GRAID’s product competes with Broadcom’s MegaRAID card and Intel’s VROC product. It was found that MegaRAID only supports four direct cabling NVMe SSDs per board, while VROC is not compatible with hardware storage, being a virtual solution.

In addition to managing local disks, the product can also operate across the entire data fabric, facilitating end-to-end integration of various data pipelines and cloud environments.

>See also: Becoming “cloud smart” — the path to accelerated digital innovation

Source: GRAID Technology

Business use cases

Key use cases for the solution, according to GRAID, include:

  • Artificial intelligence (AI) training: AI engineers can take advantage of the card to enable accelerated data processing, ranging from the usual 12 hours with more traditional solutions to as little as two. This gives staff a lot more time to focus on correcting inaccuracies before releasing the algorithm to market.
  • Automotive and aeronautics: GRAID has won business with automotive and aviation customers, helping to urgently deliver masses of transportation, distance and speed data. The compact nature of the card provides more space for businesses compared to other RAID products, resulting in cost savings.
  • Local databases: With enterprises frequently encountering latency and performance limitations when working with local databases, GRAID has helped overcome these obstacles, allowing customers to take full advantage of NVMe SSDs.
  • 4K CCTV: With 4K quality video delivery often requiring over 10GB per second of write throughput, the product is ideal for delivering processing at speed and scale.

>See also: Best practices for a modern enterprise data architecture

The start-up roadmap

Established in 2020 with an office in Ontario, California and an R&D center in Taipei, the startup completed a $15M Series A funding round in January 2022. Currently, four unique pricing plans are available for PCIe generations 3 and 4, between $2,500 and $3,995.

Going forward, GRAID aims to further develop its data center and enterprise editions. For its data center portfolio, VMWare and Kubernetes support and storage code are planned for 2023, with native cloud storage infrastructure and erasure code planned for 2024.

As for GRAID for Enterprise, an update including additional Linux distribution support and improved sequential read performance is scheduled for release later this year; PCIe Gen 5 support, erasure code and GUI compatibility for Windows is planned for 2023; and end-to-end data protection and inline drive expansion are planned for 2024.


StorPool Accelerates Software-Defined Storage Capacity — Storpool accelerates software-defined storage capacity by building large-scale storage from commodity servers.

Overcoming Chip Shortage with Extreme Data Processing — At a recent IT Press Tour in Tel Aviv, Pliops discussed how it helps customers overcome chip shortages and power-up issues. scale related to extreme data processing.


Comments are closed.