VMware recently released a product the company is calling the vSphere Storage Appliance (VSA). VSA is targeted squarely at the SMB market in which customers may not have the means to purchase expensive high-end shared storage devices, such as SANs. SANs often require specialized knowledge and skills that may be difficult or impossible for the SMB to procure.
Simply put, VSA allows customers to transform spare local storage on a vSphere host into shared storage usable by any other host. In its present release, VSA supports up to three hosts in a cluster and supports features such as local RAID within the server and can mirror data between different servers for a higher level of resiliency.
What i know about VSA 5 is there are a number of critical improvements with VMFS-5 as part of this. The most underrated is the fact there is now a unified block size of 1 MB, May be this will save a bunch of accidental formats at a lesser size.
Excellent point. With VMFS-3, a 1 MB block size, you were limited to a 256 GB volume. With previous versions of vSphere/VMFS, deciding on a block size was a critical choice. With the new unified 1 MB block size, this is no longer even a decision point and, moreover, a 1 MB block size does a lot of conserve disk space.
You may wonder how VMware managed to make such an improvement. With vSphere 5, VMware has moved to a GUID Partition Table (GPT)-based disk format as opposed to the older and less efficient Master Boot Record (MBR) format.
This change also allows vSphere to use pass-through RDM volumes in excess of 2 TB (but non-pass through RDM still have a 2 TB size limit).