With many enhancements introduced with vSphere 6.5, its time we discuss about the enhancements in the Storage. vSphere 6.5 has brought in so much granularity to the Storage management and SPBM can be made use of to the most. A lot of features make administoring and managing more simpler.
The first and foremost to discuss is the VMFS 6, its been so many days that there has been an update for the VMFS version. Though as many say that they are not directly visisble to the end user, there are performance improvements and they help in faster file creation, faster device discovery and device rescanning. VMFS 6 now supports 4K drives. Due to many on disk changes, there is no direct upgrade path from VMFS 3/5 to VMFS 6. However we have the migration path with Storage vMotion. Details will be known exactly after the GA.
Automatic Space Reclaimation/UnMap
With VMFS 6, the dead or deleted blocks on the datastore are recelaimed automatically using the crawler mechanism. Crawler mechanism reclaims the freed space within 12 hours and for faster reclaimation, we can always use the manual method using the esxclistorage vmfs unmap command.
With vSphere 6.5, it is now possible to route connectivity between iSCSI initiator and iSCSI target and different gateways can be set to different VMkernel interface. Another great feature is that it is now possible to boot ESXi from iSCSI using UEFI settings.
vSphere 6.5 allows to configure SIOC as a storage policy now and can be applied to any VM or vmdk by changing its storage policy. It is now possible to configure shares, limits and reservations on the Storage policy components. SIOC kicks in only when there is a contention as you know, so this feature is more powerfull when implemented properly.
With version 2.0, VVols can now support Array Based Replication and unlike in the earlier versions that required a VM to be on a specific datastore for replication, you can now replicate the VM as part of the VM and also as an individual VM. Virtual Volume Replication supports multiple replication targets which is a great feature for disaster recovery. For the Oracle folks out there, VVols can now has support to Oracle RAC workloads. So virtualize your DB clusters.
Automating Disaster Recovery
vSphere 6.5 offers APIs for orchestrating and automating the disaster recovery using the PowerCLI. This includes Replication Discovery, Sync Replication Groups, Recovery and Protection after a DR event.
Line of Service
vSphere 6.5 also has VASA 3.0 which introduces Line of Servcie, a group of related capabilities with a specific purpose like inspection, compression, encryption, replication and many more. This is something like a Baseline enabling to configure a service say replication and apply it to multiple storage policies at a time instead of configuring replication on each storage policy
Hope this was informative. Thanks!