After reading the content in this guide, you should understand how virtual volumes (vVols) give VMware and FlashArray more granular control than traditional datastores. You should also understand the key concepts behind that model, including vSphere API for Storage Awareness (VASA), protocol endpoints, and policy-based management.
Before You Begin
Before following the procedures in this guide, you should understand vVol architecture and have the right versions, connectivity, registrations, and policies in place. With those prerequisites met, you can use the guide to configure, protect, migrate, and report on vVol-based workloads more effectively.
We recommend that you review the Web Guide: Virtual Volumes Quick Start Guide contained in this guide.
What to Have in Place
-
A supported environment: Have compatible VMware vSphere, ESXi, and Purity//FA versions for the vVol features you want to use. This is important because capabilities such as newer VASA support or NVME over Fabrics (NVMe-oF) depend on specific releases.
-
Host and management connectivity: Have data-path connectivity from ESXi to the array and management connectivity for vCenter and compute hosts.
-
A protocol endpoint connection: You need a protocol endpoint presented to the ESXi hosts. Otherwise, the environment has management-path connectivity only and no working data path for vVol use.
-
Registered storage providers and the right tools: You should have the FlashArray//XL™ storage providers registered in vCenter and access to the interfaces used by the workflows, such as vCenter, the FlashArray UI or CLI, or the vSphere Plugin.
What To Do in Advance
-
Know which data services you want to express, such as snapshot schedules, replication settings, QoS, retention, or target arrays.
-
Before taking snapshots or performing recovery tasks, understand the snapshot behavior you want and that managed snapshots create additional array objects for each virtual disk.
-
Before migrating VMs, identify the target vVol datastore and optional storage policy. Also, understand that migration behavior differs depending on whether the source and target are on the same array.
-
Before using reporting views, know whether you want VM-level or vVol-level visibility for capacity and performance analysis.
-
Before using data-mobility workflows, confirm that the destination system can interpret the presented or copied vVol data.