VMware® Virtual SAN™ pools the compute and storage resources of VMware vSphere® hosts, providing simplicity, enterprise-class performance and significant savings.
Radically Simple Storage Management
Using Storage Policy Based Management (SPBM), a key Virtual SAN innovation, vSphere administrators create storage policies with specific capabilities relating to performance, availability and other services. At deployment time, the administrator selects the appropriate policy for each virtual machine, which SPBM then applies. This ensures that virtual machines are always assigned to the appropriate underlying storage.
Two-Click Storage Provisioning
Thanks to SPBM, provisioning is a simple matter of assigning the right storage policy to each virtual machine. To select a destination datastore, the administrator first chooses the desired level from a drop-down menu of virtual machine storage policies that have been created in advance. The datastores are then separated into compatible and incompatible categories, allowing the administrator to make the correct choice for virtual machine placement.
Seamless Integration with vSphere and VMware® vCenter™
Virtual SAN offers high availability as well as scale-out storage functionality, and it is fully integrated with core vSphere features such as High Availability, DRS and vMotion®. VSAN also facilitates quality of service by enabling and applying virtual machine storage policies that specify required performance and availability levels on a per-virtual machine basis.
High Performance with SSD Caching
Virtual SAN uses solid-state disks (SSD) as a read cache and write buffer, dramatically improving the performance of virtual machines. The read cache keeps a list of commonly accessed disk blocks to reduce I/O read latency in the event of a cache hit (that is, the disk block is in cache). The write cache behaves as a non-volatile write buffer, reducing latency for write operations as well.
In the event of a host failure, the retained copy of the in-cache data keeps it safe from corruption: The virtual machine simply reuses the replicated copy of the cache, as well as the replicated disk data.
Resiliency Against Multiple Hardware Failures
Virtual SAN leverages distributed RAID cluster architecture and cache mirroring to ensure that data is never lost in case of disk, host or network failures. Virtual SAN's SPBM feature lets administrators maximize resiliency by defining availability on a per-virtual machine basis: When setting the storage policy for each virtual machine, they can specify how many host, network or disk failures it can tolerate in a Virtual SAN cluster.
When the availability level is set for a specific virtual machine, one or more copies or "replicas" of the virtual machine storage objects are created; the number of replicas is dictated by the setting. This mechanism allows virtual machines to continue to run with a full complement of objects when there are host, network or disk failures in the cluster.
Dynamic Scaling of Performance and Capacity
Virtual SAN allows independent scaling of storage performance and capacity, as needed. You can add disks to existing hosts in the cluster at any time to increase capacity dynamically. You can also add more hosts to boost performance on the fly.
Reduced CapEx via Server Disks
Industry-standard hard disks and SSDs can be used as shared storage for virtual machines, creating a scale-out direct-attached storage system that is ideally suited for a number of common use cases. These commodity storage options are typically less expensive than specialized hardware.
High Performance, Lower TCO
Virtual SAN provides performance equivalent to traditional storage, lowering TCO by as much as 50 percent—an advantage that comes from lower capital and operating expenses, plus the ability to scale in granular increments.