VMware recently released vSphere 8.0 Update 3, packed with exciting features and enhancements. Let’s dive into the highlights. The update 3 comes out after quite a long period of time. In fact, we haven’t seen major update since U2 which has been released in August of 2023.
vSphere 8 U3 has brought some exciting new features:
Dual DPU support – Starting with ESXi 8.0 Update 3, vSphere Lifecycle manager supports 2 data processing units (DPUs) per host. These DPUs provide high availability or increased offload capacity. Dual-DPU systems can use NVIDIA or Pensando devices, and they are supported by Lenovo server designs. vSphere Lifecycle Manager will remediate both DPU ESXi versions and ensure all versions are kept at the same version.
Screenshot from VMware
CPU Enhancements:
PCIe hot plug is updated for server platforms utilizing newer generation AMD Genoa and Intel Sapphire Rapid CPUs – Starting with vSphere 8.0 Update 3, kernel hot plug is supported for newer generation CPUs such as AMD Genoa and Intel Sapphire Rapids.
Support for Intel Xeon Max Series processors with integrated High Bandwidth Memory (HBM) – vSphere 8.0 Update 3 adds support for Intel Xeon Max Series processors (formerly with code name Sapphire Rapids HBM) with 64 GB of integrated HBM, aimed to enhance performance for workloads like high performance computing apps, artificial intelligence (AI), and machine learning (ML).
CPU C-State Power virtualization – With vSphere 8.0 Update 3, for use cases such as Virtualized Radio Access Network (vRAN) workloads, you can configure and control the C-State power of the physical CPUs dedicated to vRAN VMs from the vSphere Client.
vSphere IaaS Control Plane on vSAN Stretched Clusters – You can now run the vSphere IaaS control plane (formerly known as vSphere with Tanzu) on vSAN stretched clusters.
Note that this capability is available only for greenfield installations of the IaaS control plane on stretched vSAN clusters.
Virtual Machine Management – Introducing a new Virtual Machine Compute Policy for best effort evacuation during host maintenance mode.
When a host enters maintenance mode, VMs are shut down. If shutdown fails, they are powered off. The new policy attempts to power them on periodically on the best available ESXi host.
Embedded vSphere Cluster Service (vCLS) – vSphere 8.0 Update 3 redesigns vCLS to Embedded vCLS, leveraging vSphere Pod technology.
What is Embedded vCLS?
Embedded vCLS is a feature introduced in vSphere 8.0 Update 3.
It uses CRX runtime technology, initially developed for the vSphere IaaS control plane (formerly known as vSphere with Tanzu).
Unlike previous versions, Embedded vCLS VMs have no storage footprint and run entirely in host memory.
These VMs are automatically activated on vSphere clusters containing at least one ESXi 8 Update 3 host and are managed by vCenter 8 Update 3.
Key Changes and Benefits:
No OVF Deployment: Unlike before, there’s no OVF deployment pushed from vCenter, and EAM (ESX Agent Manager) is no longer involved.
No vMotion Support – vSphere vMotion and Storage vMotion are not supported when using Embedded vCLS. This is due to the changes in VM structure and the absence of a storage footprint.
Reduced VM Count – The number of vCLS VMs per cluster has been reduced from up to three to two VMs when using Embedded vCLS. A single-node cluster uses one VM, while clusters with two or more hosts use two VMs.
Automated Maintenance Mode – When placing a host running an Embedded vCLS VM into maintenance mode, the host automatically spins down the VM, and another host in the cluster spins up a new one.
Remaining Aspects:
System Managed – Embedded vCLS VMs are system-managed, allowing only power-off and hard stop operations for troubleshooting.
Required for DRS and HA – Embedded vCLS is essential for vSphere DRS and vSphere HA to operate optimally.
Mixed Version Clusters – Embedded vCLS activates automatically when at least one host is updated to ESXi 8 Update 3. Always update vCenter to 8 Update 3 before updating hosts1.
Remember, Embedded vCLS ensures cluster services’ availability even if vCenter Server becomes unavailable
Reduced Time to Inflate VMFS Disks – A new VMFS API allows you to inflate a thin-provisioned disk to eagerzeroedthick (EZT) while the disk is in use, up to 10 times faster than previous methods.
vSphere Live Patch is a powerful feature introduced in vSphere 8.0 Update 3 that significantly reduces the impact of patching on your virtual infrastructure.
Here’s what you need to know:
What is vSphere Live Patch?
With vSphere Live Patch, you can address critical bugs in the virtual machine execution environment (vmx) without the need to reboot or evacuate the entire host.
It allows vSphere clusters to be patched live, meaning you don’t have to migrate workloads off the target hosts or put the hosts into full maintenance mode.
The patch is applied while workloads continue to run, minimizing disruption. Some Virtual machines are fast-suspend-resumed (FSR) as part of the host remediation process.
How It Works:
When applying a patch, the ESXi host enters partial maintenance mode. Existing VMs continue to run, but new VMs cannot be created on the host or migrated to/from it.
Screenshot from VMware.
A new revision of the target patch components is mounted alongside the current version.
The new revision files and processes are patched, and virtual machines undergo a fast-suspend-resume process to consume the patched revision.
Not All Patches Are Equal:
Initially, vSphere Live Patch targets patches for the virtual machine execution component of ESXi. These patches can be applied live.
Patches that change other areas of ESXi (e.g., VMkernel patches) still follow the existing patching workflow, requiring maintenance mode and VM evacuation.
Each Live Patch is compatible with specific ESXi versions, so check compatibility before applying.
Remember that VMs with vSphere Fault Tolerance or using Direct Path I/O must be manually remediated.
vCenter Reduced Downtime – New Update
We know that vCenter Server HA configuration can help you when you lose vCenter server VM. U3 and vCenter Reduced Downtime update supports the existing innovations, as well as brings some enhancements such as Automatic switchover which is basically allowing you to control when you switch from one vCenter server to another.
Screenshot from VMware showing a new option to automatically complete the switchover, or you can manually initiate the switchover phase
New topology support in vSphere 8 U3 allows you chose topology that fits best for your organization:
Self-managed – vCenter VM is managed by itself.
Non self-managed – vCenter VM is managed by a different vCenter.
Note: The managing vCenter should be version 7.0 or later
Enhanced Linked Mode – Two or more vCenter instances participating in the same SSO domain.
Note: Do not update multiple vCenter instances in parallel when participating in enhanced linked mode
vCenter HA – vCenter instances configured for vCenter High Availability.
vCenter HA is automatically removed before the update begins and automatically reactivated once the update has completed.
vCenter HA self-managed and non self-managed topologies are supported for vCenter Reduced Downtime Update.
vCenter HA instances must be at least vCenter 8 Update 2 to allow reduced downtime update orchestrate the operation.
Source: VMware
