Latest news
HardwareQSANQSAN 3U60 NVMe All-Flash Array

QSAN to Showcase 3U60 NVMe Storage for On-Premises AI at SC26

QSAN will showcase its 3U60 All-Flash Array with JetStor at SC26, taking place November 15–20. Designed to house up to 60 NVMe SSDs in a 3U chassis, the system targets on-premises AI and HPC workloads.

Server rack with a high-density NVMe storage system in a modern data centerAI image
Representative image, generated with AI.Image: 3dsınıfı / FCA AI

In brief

  1. QSAN will showcase the 3U60 NVMe All-Flash Array at SC26, taking place November 15–20, 2026.
  2. The system is designed to house up to 60 NVMe SSDs in a 3U chassis.
  3. QSAN says its OCP card-based architecture aims to deliver storage density and infrastructure flexibility.
  4. The company says selected NVMe platforms support NVIDIA GPUDirect Storage.

QSAN Technology announced on November 11 that it will showcase its 3U60 NVMe All-Flash Array storage system with JetStor at SC26, which will take place November 15–20, 2026, at McCormick Place in Chicago. Positioned for on-premises AI infrastructure, the system is designed to house up to 60 NVMe SSDs in a 3U chassis. QSAN and its U.S. distribution partner, JetStor, will be at booth 7207.

OCP card-based design takes center stage

One of the highlights of QSAN’s announcement is the OCP card-based architecture used in the 3U60. The company says this approach is intended to increase storage density in limited rack space and provide flexibility in hardware configuration. The manufacturer describes the system as suited to the low-latency, high-throughput access requirements of AI data pipelines, large datasets, and high-performance computing (HPC) applications.

The “60” refers to the number of NVMe SSDs the chassis can house; the announcement does not specify total storage capacity. QSAN also did not share detailed technical specifications, pricing, or compatibility information for the system’s software and hardware. As a result, the SSD count alone does not indicate how much data it can store or how it will perform on specific workloads.

AI data flows and GDS support

QSAN positions its enterprise NVMe portfolio as supporting a data lifecycle that spans data ingestion and preparation, model training and inference, and long-term storage. The company also lists RAG applications, vector databases, and HPC workloads among its use cases.

The manufacturer says selected NVMe platforms support NVIDIA GPUDirect Storage (GDS), which is intended to make data transfers more efficient in GPU-accelerated workloads. However, the announcement does not explicitly state whether the 3U60 model being shown at SC26 supports this feature. QSAN also mentions features for protecting sensitive data and supporting business continuity, but does not explain how they apply to the 3U60.

Implications for architecture and visualization workflows

The 3U60 is not a modeling tool, render engine, or workstation; it is an enterprise storage system announced for data centers. For architecture and visualization firms, a potential use case could be storing and accessing large datasets used in in-house AI or HPC workloads. However, QSAN has not announced any dedicated integration with architectural software, render workflows, or specific project types.

Teams evaluating the product should first review their data volumes and workflows, then confirm with the manufacturer whether the 3U60 is compatible with their existing infrastructure, supports GDS, and meets their detailed technical requirements. Conversations with QSAN and JetStor representatives at SC26 could provide an opportunity to get answers. With neither pricing nor capacity information announced, it is not yet possible to estimate the purchase cost or how the system might fit an office-scale setup.

Sources

1 source
H
HPCwirehpcwire.com/off-the-wire/qsan-and-jetstor-to-showcase-3u60-nvme-all-flash-storage-for-on-prem-ai-at-sc26
Summary

Source texts are not republished; short quotes are marked, everything else is our own summary and commentary.

3dsınıfı’s take
3dEditor’s assessment

The potential value of the 3U60 for architecture and visualization firms lies less in a direct promise of faster rendering and more in its focus on the storage layer for in-house AI and HPC work. Its high-density storage approach may appeal to teams working with large datasets, but the announcement provides no architecture-specific software integration or measured performance results.

For offices in Turkey, compatibility, actual storage needs, and total cost should all be considered before making an investment decision. With pricing and technical details still undisclosed, the practical benefit the 3U60 could bring to existing infrastructure remains unclear.

Frequently asked questions

When will the QSAN 3U60 NVMe All-Flash Array be showcased?

QSAN will showcase the system with JetStor at SC26 in Chicago, taking place November 15–20, 2026.

How many NVMe SSDs can the QSAN 3U60 house?

According to QSAN’s announcement, the 3U chassis can house up to 60 NVMe SSDs. Total storage capacity was not disclosed.

Does the 3U60 support NVIDIA GPUDirect Storage?

QSAN says selected NVMe platforms support GDS, but the announcement does not clarify whether the 3U60 supports the feature.

Comments and the forum are in Turkish.Join the discussion
+

Related news