VMware vSAN 8 U2 Announced at Explore 2023

At VMware Explore US 2023 in Las Vegas, VMware announces vSAN 8 U2!.

vSAN 8 U2 — What’s new #

Nothing really big and revolutionary in the core vSAN 8 U2 announcement, besides vSAN Max that is. It is nice to see that native file services in vSAN ESA are now in feature parity with vSAN OSA, and that the configuration of 2-Node and Stretched Clusters as been simplified. Other than that, everything else seems to be performance improvements, simplified management and scalability improvements but no stand-out new features in the core offering itself.

Here’s a quick rundown of the improvements in vSAN 8 U2 #

vSAN ESA now supports native file services #

• Supports NFS and SMB protocols for traditional and cloud native clients
• Feature parity with vSAN file services in Original Storage Architecture

Adaptive Write Path Optimizations in vSAN ESA #

• Use of more in-memory I/O banks to process incoming writes more quickly
• Dynamic and scalable based on the demands of the workloads
• Improves aggregate and single object (VMDK) performance in vSAN ESA
  • Higher IOPS
  • Higher throughput — Lower latency

Improved Performance for Disaggregated Environments #

• Dynamically uses optimal write path for guest VM writes

  • Default write path
  • Large I/O write path

• Includes other optimizations made to Adaptive Write Path in vSAN 8 U2

• Better performance in disaggregated environments

  • Up to 200% higher write throughput
  • Upto 70% reduction in latency

Higher Throughput and IOPs for I/O Intensive Workloads #

Improved parallelism with vSAN ESA’s object manager client

• Delivers higher IOPS and throughput for single object, I/O intensive workloads
• Compliments I/O processing improvements in vSAN 8 U1

Reduces potential bottleneck conditions high in the storage stack Most beneficial for:

• Mission critical applications, databases, etc.
• Resource intensive VMs
• Single object (VMDK) performance evaluations

Support for up to 500 VMs per host in clusters running vSAN ESA #

• 2.5x the density of earlier versions
• Allows for new high-density cluster designs
  • Exploit benefits of vSAN ESA
    • Capacity efficiencies using erasure coding
    • Minimal resource overhead
    • High performance

New ReadyNode profile and support for Read-Intensive devices for vSAN ESA #

New ReadyNode profile ideal for small data centers and edge environments

Dramatically lower TCO while benefiting from capabilities in ESA

• New snapshots

• Smaller failure domains

• Improved efficiency

• New support for Read- Intensive NVMe devices

  • Available in some ReadyNode profiles
  • Supported on vSAN 8 U1 and later

Auto-Policy Management improvements in vSAN ESA #

• Improves Auto-Policy Management feature introduced in vSAN 8 U1
• Single click “Update Cluster DS Policy” to remediate change
• Updates prescribed, cluster- specific storage policy if number of hosts in cluster change
• Streamlines handling of ongoing changes to cluster reduces manual intervention

Capacity overhead reporting improvements in vSAN ESA #

• Provides “ESA object overhead” in cluster capacity usage breakdown view
• Represents capacity consumed for vSAN’s log- structured filesystem (LFS) for all object and replica data

vSAN ESA Prescriptive Disk Claim #

• Uses new outcome oriented “desired state” disk claiming model
• Define disk claim criteria for hosts in cluster, and let vSAN take care of the rest
• Provides consistency to disk claiming process for initial deployment and cluster expansion
• Continuous checks of compliance to desired state
• Available using API and CLI

Support for key expiration with Data-at-Rest Encryption in vSAN OSA and ESA #

Support of key expiration standard for KMIP-based Key Management Servers Integrated with Skyline Health for vSAN Triggered health finding will provide:

• Days remaining on valid keys
• Convenient way to perform a shallow rekey

Skyline Health for vSAN remediation enhancements #

• Health finding recommendations customized to specific versions of vSAN
• Improved information on risks of not remediating triggered finding
• More prescriptive options to fix identified issue
  • Default
  • Alternative
  • New endurance monitoring of NVMe devices in vSAN ESA

Top Contributors enhancements in vSAN OSA and ESA #

• Find performance hot spots over a customizable time period when diagnosing performance issues
• Multiple source types available for analysis
  • VMs
  • Hosts(frontend)
  • Hosts(backend)
  • Host disks
• Transpose other top contributors in same performance view to understand correlations

I/O Trip Analyzer for 2-Node and Stretched Clusters in vSAN OSA and ESA #

• Capture diagnostics data on workloads running in a vSAN stretched cluster or 2-node cluster
• Easily identify primary source of latency
  • Outstanding I/Os of object
  • Network link across sites
  • Network links within sites
  • Host storage devices
  • Supports all data placement schemes
    • RAID-1
    • RAID-5
    • RAID-6

Witness Traffic Separation and other configuration improvements in vSAN OSA and ESA #

• Easily configure Witness Traffic Separation in UI
• Removes need for CLI commands to properly tag witness traffic
• Applies to both 2-Node and stretched clusters
• New support for “medium” sized witness host appliance in vSAN ESA
• New support in vLCM to manage lifecycle of all witness host appliance types

This is a post in the VMware Explore US 2023 series. Posts in this series:

Related Posts

• VMware vSAN 8 U2 to the Max — Published 22 Aug, 2023
• vSAN 8 ESA VMware Compatibility Guidance — Published 7 Dec, 2022
• VMware vSAN 8 ESA in my Homelab — Published 24 Nov, 2022
• VMware vSAN 8 Announced — New Architecture Model! — Published 30 Aug, 2022
• VMware vSphere 8 — The Enterprise Workload Platform Announced! — Published 30 Aug, 2022