Menu

VMware Disaggregates HCI with vSAN Max

VMware Disaggregates HCI with vSAN Max

The News: On August 22, VMware announced vSAN Max, a petabyte-scale disaggregated storage architecture for VMware workloads. You can read the entire Press Release here.

VMware Disaggregates HCI With vSAN Max

Analyst Take: vSAN Max is a storage-only cluster for any VMware vSphere cluster. The offering is an extension of VMware’s vSAN hyperconverged infrastructure (HCI) software and is licensed separately from vSAN. Using vSAN Max, customers can scale and license the compute and storage nodes separately – something not possible with traditional HCI.

What Is vSAN Max?

vSAN Max, unveiled at VMware Explore 2023, is VMware’s second move toward disaggregated scaling. VMware first introduced disaggregated vSAN through HCI Mesh in 2020. HCI Mesh allows customers to share unused capacity across vSAN HCI clusters. HCI Mesh compute clusters can use storage from a remote vSAN cluster, and the compute clusters do not need vSAN licenses. That means customers no longer must buy nodes with compute and storage if they only need storage capacity but have enough compute.

vSAN Max uses the vSAN Express Storage Architecture (ESA) introduced in vSAN 8 at VMware Explore 2022, plus capabilities of HCI Mesh.

vSAN Max scales to 24 data nodes per cluster, with each node scaling from 200 TiB to 360 Tib for a maximum of 8.6 PB per cluster. Each cluster can support more than 100 vSphere nodes, and reach 3.6 million IOPS. Max customers will manage clusters and provision storage through vCenter Server, just as they manage traditional vSAN clusters.

Most of the scaling benefits of vSAN Max were available through HCI Mesh. There are some differences, though. vSAN Max can be created from a stretched cluster spanning data centers for availability, and HCI Mesh did not support that.

Another difference is in the architecture. VMware gives vSAN customers a choice of ESA or the original storage architecture, but vSAN Max requires ESA. The major difference between the architectures is that ESA was designed for storage devices that did not exist when vSAN first launched, such as high-performance NVMe-based flash devices.

Max uses vSAN ESA high-density ESA Ready Nodes, along with ESA’s snapshots, communications between nodes, and storage policies. Customers can choose between a single-site or stretched cluster topology.

vSAN Max is expected to be available by the first quarter of 2024. VMware plans to sell it as a subscription, licensed per TiB.

If It Is Disaggregated, Is It Still HCI?

The major limitation of HCI from the start has been that because storage and compute are integrated inside each node, the infrastructure does not support independent scaling of those resources. If a customer needs only compute or storage, they must add both with a traditional HCI node.

HCI became a multi-billion dollar product category on the strengths of products such as vSAN, Nutanix Cloud Platform, and Dell VxRail (which uses vSAN software) that aggregated storage and compute. Other products such as HPE dHCI and NetApp HCI (now discontinued) followed with the intention of disaggregating resources again for more efficient resource scaling. The Nutanix Cloud Platform and Dell VxRail HCI systems now also support compute-only nodes for independent scaling of compute and storage.

VMware calls vSAN Max a new approach, but it is only new to VMware, and not even completely new. It has similarities to the traditional networked storage architecture before HCI, and the disaggregated HCI products that followed. In this case, it is vSAN Max storage nodes connected to VMware vSphere compute nodes. However, vSAN Max is fully distributed with no need for paired controllers with clustered file systems, or iSCSI or NFS storage protocols. It is a unified block, file, and object storage system.

Why Disaggregation?

VMware lists popular vSAN use cases as:

  • Cloud native applications
  • Business-critical applications
  • More foundational applications that companies need to run their business such as relational databases, VDI workloads

Customers like HCI’s simpler management from the combination of compute and storage, and easier storage provisioning. But HCI’s inability to scale compute and storage independently could be problematic when using more storage-intensive applications on vSAN. For instance, OLTP workloads and ERP databases often require storage capacity to scale higher than compute resources. That prompted VMware to find ways to scale storage without scaling compute.

I think vSAN Max will appeal to new VMware storage customers, but it is unclear if existing customers on the vSAN original storage architecture will switch over the ESA.

Disclosure: The Futurum Group is a research and advisory firm that engages or has engaged in research, analysis, and advisory services with many technology companies, including those mentioned in this article. The author does not hold any equity positions with any company mentioned in this article.

Analysis and opinions expressed herein are specific to the analyst individually and data and other information that might have been provided for validation, not those of The Futurum Group as a whole.

Other insights from The Futurum Group:

VMware Doubles Down on Cross-Cloud Services

Breaking Down the Myths on Broadcom-VMware

Talking Broadcom, China, MSFT, Adobe, Amazon, IBM & Google

Author Information

Dave focuses on the rapidly evolving integrated infrastructure and cloud storage markets.

Related Insights
Coherent Q2 FY 2026 AI Datacenter Demand Lifts Revenue and Margins
February 6, 2026

Coherent Q2 FY 2026: AI Datacenter Demand Lifts Revenue and Margins

Futurum Research analyzes Coherent’s Q2 FY 2026 results, highlighting AI datacenter optics demand, 6-inch indium phosphide capacity expansion, and growing OCS/CPO traction supporting margin expansion into FY 2027....
Arm Q3 FY 2026 Earnings Highlight AI-Driven Royalty Momentum
February 6, 2026

Arm Q3 FY 2026 Earnings Highlight AI-Driven Royalty Momentum

Futurum Research analyzes Arm’s Q3 FY 2026 results, highlighting CPU-led AI inference momentum, CSS-driven royalty leverage, and diversification across data center, edge, and automotive, with guidance pointing to continued growth....
Qualcomm Q1 FY 2026 Earnings Record Revenue, Memory Headwinds
February 6, 2026

Qualcomm Q1 FY 2026 Earnings: Record Revenue, Memory Headwinds

Futurum Research analyzes Qualcomm’s Q1 FY 2026 earnings, highlighting AI-native device momentum, Snapdragon X PCs, and automotive SDV traction amid near-term handset build constraints from industry-wide memory tightness....
Alphabet Q4 FY 2025 Highlights Cloud Acceleration and Enterprise AI Momentum
February 6, 2026

Alphabet Q4 FY 2025 Highlights Cloud Acceleration and Enterprise AI Momentum

Nick Patience, VP and AI Practice Lead at Futurum analyzes Alphabet’s Q4 FY 2025 results, highlighting AI-driven momentum across Cloud and Search, Gemini scale, and 2026 capex priorities to expand...
Amazon CES 2026 Do Ring, Fire TV, and Alexa+ Add Up to One Strategy
February 5, 2026

Amazon CES 2026: Do Ring, Fire TV, and Alexa+ Add Up to One Strategy?

Olivier Blanchard, Research Director at The Futurum Group, examines Amazon’s CES 2026 announcements across Ring, Fire TV, and Alexa+, focusing on AI-powered security, faster interfaces, and expanded assistant access across...
NXP Q4 FY 2025: Auto Stabilises, Edge AI Platforms Gain Traction
February 5, 2026

NXP Q4 FY 2025: Auto Stabilises, Edge AI Platforms Gain Traction

Futurum Research analyzes NXP’s Q4 FY 2025 earnings, highlighting SDV design wins, edge AI platform traction, and portfolio focus, with guidance pointing to steady margins and disciplined channel management into...

Book a Demo

Newsletter Sign-up Form

Get important insights straight to your inbox, receive first looks at eBooks, exclusive event invitations, custom content, and more. We promise not to spam you or sell your name to anyone. You can always unsubscribe at any time.

All fields are required






Thank you, we received your request, a member of our team will be in contact with you.