Close Menu
DPC Virtual Tips
    Read More

    How to Troubleshoot Packet Drops on an ESXi Host

    August 16, 2026

    SlurmDBD Is Down: What Continues Working and What Does Not

    August 15, 2026

    How to Investigate Jobs Stuck in COMPLETING State on Slurm

    August 14, 2026
    • Home
    • About Us
    • Contact
    • Cookie Policy
    • Comment Policy
    • Privacy Policy
    • Terms of Use
    • Disclaimer
    Tuesday, August 25
    DPC Virtual Tips
    • Home
    • Operating Systems
    • PowerFlex
    • HPC
    • Virtualization
    • About the Author
    • About Us
    • Contact
    DPC Virtual Tips
    Home » What is a vSAN ESA Storage Pool?
    Virtualization

    What is a vSAN ESA Storage Pool?

    DaniloBy DaniloApril 18, 2024Updated:July 31, 2026No Comments4 Mins Read
    Facebook Twitter Pinterest LinkedIn Tumblr Email
    vSAN ESA storage pool
    Share
    Facebook Twitter LinkedIn Pinterest Email

    Understand the vSAN ESA storage pool: how it groups NVMe SSDs, differences from OSA disk groups, and current ReadyNodes support.

    First and Foremost, What is the vSAN ESA?

    vSAN ESA is an acronym for Express Storage Architecture.

    vSAN Express Storage Architecture, or ESA, is a new vSAN architecture to process and store data. It is an optional or alternative architecture for vSAN. You can already use the Original Storage Architecture (OSA).

    vSAN ESA is optimized to exploit the full potential of the latest hardware and unlock new capabilities. At this moment, vSAN ESA is only supported on ReadyNodes servers.

    Before talking about some aspects of the storage pool in the vSAN ESA, let’s remember how the vSAN disk group works in the vSAN OSA.

    vSAN OSA – Hybrid Disk Group

    vSAN hybrid disk group is based on a two-tier architecture: cache tier and capacity tier.

    Cache Tier: SSD disks

    • 70% used for reads
    • 30% used for writes

    Capacity Tier: HDD disks

    All reads are served from the cache tier and all write operations are acknowledged when they arrive at the cache tier.

    Generally, when the write buffer is almost used (~ 30% of usage), the destage process is started moving the data from the cache tier to the capacity tier.

    This type of vSAN cluster (Hybrid) requires at least 1 Gbps for network bandwidth (it means that each ESXi host needs to have at least 1 Gbps network connection for the vSAN network).

    In the following figure, we can see how a vSAN hybrid disk group works:

    Each ESXi host can have a maximum of 5 disk groups. Each disk group has 1 disk for cache and 1 to 7 disks for capacity.

    vSAN OSA – All-Flash Disk Group

    In a vSAN All-Flash cluster, we already have the two-tier architecture. However, both tiers are made by SSD devices (SSD or NVMe):

    • Cache Tier: SSD
    • Capacity Tier: SSD

    With an all-flash disk group, all reads are served directly from the capacity tier. The cache tier is allocated 100% for write operations.

    This type of vSAN cluster (All-Flash) requires at least 10 Gbps for network bandwidth (each ESXi host must have at least 10 Gbps connection for the vSAN network).

    In the following figure, we can see how a vSAN all-flash disk group works:

    Each ESXi host can have a maximum of 5 disk groups. Each disk group has 1 disk for cache and 1 to 7 disks for capacity.

    vSAN ESA – Storage Pool

    With the vSAN ESA, we did not have the vSAN disk group. All disks are grouped in an entity called a storage pool.

    Only all-flash devices (NVMe SSD devices) are supported. The recommendation is to use NVMe TLC devices. In the context of NVMe disks, TLC stands for Triple-Level Cell:

    • TLC SSDs write three bits to each cell, allowing for higher storage capacity compared to Single-Level Cell (SLC) and Multi-Level Cell (MLC) SSDs;
    • Due to their higher storage density, TLC SSDs generally have a lower cost per bit, making them more affordable than SLC and MLC SSDs;
    • When combined with the NVMe protocol, TLC SSDs can offer fast data transfer speeds and reduced latency compared to traditional SSDs using the SATA interface.

    Since the storage pool does not have two tiers, each disk in the storage pool is independent.

    The number of disk slots defines the maximum number of disks (we have ReadyNodes profiles to select the type of hardware supported by vSAN ESA).

    In the following figure we can see how a storage pool works:

    Storage Pool Failure Handling

    When a disk failure occurs, vSAN ESA uses the defined storage policy to rebuild the missing components on another NVMe SSD disk (failed disk components are rebuilt on other disks automatically, based on their vSAN Storage Policy).

    As each disk is independent in a vSAN ESA storage pool, a failure of one disk does not impact other disks in the cluster).

    To Wrapping This Up

    As we can see, with vSAN ESA we do not have the disk group anymore. However, with the vSAN OSA, the disk group concept continues working normally. With vSAN ESA the storage pool is introduced, only supporting NMVe devices. Because of that, it’s important to understand all vSAN ESA hardware requirements before building your cluster – and not forget – vSAN ESA is currently supported only in vSAN ReadyNodes.

    Share. Facebook Twitter Pinterest LinkedIn Tumblr Email
    Previous ArticleComparing vSAN Architectures OSA x ESA
    Next Article Working with a vSphere Resource Pool
    Danilo

    Infrastructure Engineer with experience in Virtualization, Linux, Windows Server and learning automation using Python. DPC Virtual Tips was created to share practical tutorials, lab experiences and troubleshooting guides focused on enterprise infrastructure technologies.

    Related Posts

    How to Troubleshoot Packet Drops on an ESXi Host

    August 16, 2026

    VM Replication with vSphere Replication and Site Recovery Manager

    June 16, 2025

    Changing DNS Settings on ESXi Hosts

    May 20, 2025

    Comments are closed.

    Search
    Categories
    • HPC (12)
    • Operating Systems (85)
    • PowerFlex (22)
    • Virtualization (130)
    Read More
    Virtualization

    How to Troubleshoot Packet Drops on an ESXi Host

    By DaniloAugust 16, 20260
    HPC

    SlurmDBD Is Down: What Continues Working and What Does Not

    By DaniloAugust 15, 20260
    HPC

    How to Investigate Jobs Stuck in COMPLETING State on Slurm

    By DaniloAugust 14, 20260
    Operating Systems

    Linux Server Has Free Memory but Is Swapping: Why?

    By DaniloAugust 13, 20260
    Operating Systems

    How to Determine Whether Packet Loss Is Local or Network Related on Linux

    By DaniloAugust 12, 20260
    Latest Posts

    How to Troubleshoot Packet Drops on an ESXi Host

    August 16, 2026

    SlurmDBD Is Down: What Continues Working and What Does Not

    August 15, 2026

    How to Investigate Jobs Stuck in COMPLETING State on Slurm

    August 14, 2026
    Images from Gallery
    hpc main commands
    linux commands
    install rock linux
    lustre fs
    shell scripting
    vSAN Trace Files
    Categories
    • HPC
    • Operating Systems
    • PowerFlex
    • Virtualization
    • Home
    • About Us
    • Contact
    • Cookie Policy
    • Comment Policy
    • Privacy Policy
    • Terms of Use
    • Disclaimer
    Copyright © 2026, DPC Virtual Tips. All rights reserved.

    Type above and press Enter to search. Press Esc to cancel.

    We use cookies to ensure your best experience on our website. If you continue using our website, we'll assume you agree to our cookie policy