Pure Storage Unveils High-density QLC Flash Advancements

AI and other new workloads are increasing storage demand, requiring high performance per petabyte in hot/warm tiers. Hard disks can’t keep up, and triple-level cell (TLC) flash is costly and power-intensive. Quad-level cell (QLC) flash offers the ideal balance of cost, performance, space, and power efficiency for hyperscale storage.

Pure Storage is redefining how hyperscalers interact with and manage flash storage by developing a hyperscale architecture to leverage host-based flash management. This shift signifies a leap forward, offering a more efficient, scalable, and responsive storage infrastructure.

At SNIA Developer Conference 2025, Pure Storage unveiled high-density QLC flash and an innovative approach to host-based flash management, based on Purity software. This purpose-built, software-defined architecture delivers industry-leading performance density, power, and space efficiency in a cost-effective manner.

The Era of High-density QLC Flash in Hyperscale Storage

For years, QLC flash has promised high-density, cost-effective storage, but challenges in endurance and performance have slowed its widespread adoption. Our keynote presentation, “How to Build Exascale Storage with QLC,” showcased how to overcome these hurdles through:

• Advanced error correction and data protection schemes: Ensure longevity and integrity of QLC-based systems
• Intelligent wear-leveling mechanisms: Extend flash life span
• Performance optimization for mixed workloads: Delivers consistent performance even under demanding, mixed read/write workloads with tunable configurations such as indirection unit (IU) sizes

The Power of Host-based Flash Management 

In the “Optimizing Hyperscale Flash Storage” session, Pure Storage highlighted the benefits of a host-based flash translation layer (FTL). Traditionally, the FTL on the flash device introduces overhead and can limit the flexibility, durability, and efficiency of the flash. The session challenged this long-held paradigm, highlighting the benefits of a host-based FTL, including:

• Greater flexibility and scalability: Scaling FTL operations with compute resources offers inherent advantages, enabling agile and adaptable storage solutions compared to traditional SSDs.
• Reduced latency and increased throughput: Offloading FTL operations to the host significantly reduces latency and delivers substantial gains in I/O throughput by streamlining data paths and minimizing bottlenecks.
• Enhanced resource utilization: Host-based FTL enables intelligent and dynamic resource allocation, leading to greater efficiency and better overall system performance.

Hear from Pure Storage Experts at OCP Global Summit 2025

Join us at the 2025 OCP Global Summit in San Jose, California, on October 13-16. Learn more about the Pure Storage hyperscale architecture in action and explore how you can help shape open specifications and deployment models to accelerate adoption.

Don’t miss these sessions:

Deploying QLC at Scale: Balancing Density, Power, Performance, and Reliability

Ben Di Qual, Senior Director Product Management, will share how Pure Storage QLC flash scales and how Pure Storage host-based flash media management (FMM) boosts endurance, reliability, and optimizes memory utilization.

Panel: EDSFF New and Upcoming Updates

Peter Choi, Engineering Manager, will discuss the Pure Storage-backed SNIA EDSFF E2 form factor, which delivers petabyte-per-node and exabyte-per-row efficiency.

Register now on the Open Compute Project website.