In this post we will focus on 3 key product enhancements around vSAN.

Simpler Lifecycle Management

Increase reliability and reduce number of tools

In vSphere 6.x hosts are individually managed with VMware Update Manager (VUM), but vSAN is a cluster based solution which is not ideal and can create inconsistencies.

vSphere 7 is introducing an entirely new solution at the cluster level to unify software and firmware management.

• 
• This new approach is focused around this desired-state model for all lifecycle operations
  • Monitors compliance “drift” in real time
  • Provides then the ability to remediate back to the desired state
• Built to manage server stack in cluster
  • Hypervisor
  • Drivers
  • Firmware
• Modular framework supports vendor firmware plugins which allows their own firmware and respective drivers to be integrated with an image that is applied to hosts.
  • Dell
  • HPE

Native File Services

Now we talking! vSAN 7 introduces a fully integrated file service that is built right into the Hypervisor and managed through vCenter Server.

• 
• Provision vSAN cluster capacity for file shares Supports NFS v4.1 & v3
• Supports quotas for file shares
• Suited for Cloud Native & traditional workloads on vSAN
  • I don’t think this capability is looking to replace large scale filers, but more looking to solve the specific use cases within that particular cluster.
• Works with common vSAN/vSphere features

There are many use cases for both traditional VMs as well for cloud native applications. Let’s look at the latter.

Extension and integration K8s running on vSphere and vSAN

• 
• Native files services will offer file-based persistent storage
• vSAN also provides persistent block storage through SPBM for vSAN and vVols which is associated to a Storage class in Kubernetes
• Persistent volume encryption and snapshot support
• Volume resizing support
• Support for some different tooling options
  • Wavefront
  • Prometheus
  • vROps

Enhanced Cloud Native Storage

 

Stretched Cluster and 2-Node Topologies

Support for overriding default gateway

• 
• This will simplify deployments for routed topologies, greenfield builds as well as cluster transitions. No more logging directly into each ESXi host to set statics routes!

Immediate repair operation after a witness host appliance is replaced

• 
• “Replace Witness” workflow will now invoke an immediate repair to regain compliance
  • In vSAN 6.x it would register the new witness host only after the timeout window
• Applies to stretched cluster and 2-node topologies
• Minimizes interruption with site-level protection

Improved VM placement intelligence for stretched clusters

• 
• Prioritizes I/O read locality over any VM site affinity rules
• Instructs DRS not to migrate VMs to desired site until re-syncs complete
• Reduces I/O across ISL in recovery conditions
  • Improve VM read performance
  • Free up ISL for resyncs to regain compliance

Intelligent capacity management for stretched cluster

Site capacity can be become imbalanced due to different factors like for instance storage policies with site affinity. This can create critical threshold for a specific object. This new feature will improve VM uptime.

• 
• Prevents capacity imbalance to impact VM uptime in stretched clusters
• Redirects active I/O to the site with available capacity
• Allows for VMs to continue non-disruptively in capacity
• Assumes rebalancing within site has taken place

Operations and Management Enhancements

 

VMware Skyline integration with vSphere Health and vSAN Health

• 

Improved Consistency in VM Capacity Reporting

• 
• Consistent VM-level capacity usage across vCenter UI and APIs
• Accurately accounts for used capacity of:
  • Thin provisioned VMs
  • Swap Object
  • Namespace Object
• Reduces confusion on capacity consumed by a VM

Easily View Memory Consumption for vSAN services

This will help you better understand the memory consumption as a result of running vSAN

• 
• New vSAN memory metric available in API and UI
• Time-based memory consumption details per host
• View consumption as a result of hardware and software configuration changes
• Adding devices or disk groups
• Enabling/disabling data services

Improved Awareness and visibility of vSphere Replication Data in vSAN cluster

In vSAN 6.x if you were using vSphere replication and you were sending it to a target that was another vSAN cluster it would just come up as an unknown object, even though it is available.

• 
• Easily identify objects created by and used with vSphere Replication
• New vSphere Replication object identity type in “Virtual Objects” listing
• New vSphere Replication categories for cluster-level capacity view

Hardware, Resilience and Usage Enhancements

Increase Efficiency with the Latest Hardware

• 
• Support for much larger storage devices
  • 32TB physical capacity
  • 1PB in logical capacity (DD&C)
• Potential for Improved deduplication ratios with larger devices
• Designed to minimize data movement for support of new devices

Flexible Serviceability for NVMe Devices Improves Uptime

Native support for planned and unplanned maintenance with NVMe hot plug

• 
• Hot plug support for both vSphere and vSAN
• Better TCO through RASM
  • Reliability
  • Availability
  • Serviceability
  • Manageability
• Minimize host restarts
• Reduces complexity of steps to service systems
• Select OEM platforms only

Improved Flexibility for Applications using Shared Disks on vSAN

This improvement are for customers that run Oracle RAC on vSAN

• 
• Removal of Eager Zero Thick (EZT) requirement for shared disk in vSAN
• Applies to all shared virtual disks using multi-writer flag (MWF)
• Object Space Reservation (OSR) storage policy rule set to 100 is no longer necessary for shared disks

 

(All images on this page courtesy VMware)