We are thrilled to finally be sharing all the great features coming in Virtual SAN 6.1 and hope you read all about the updates here; however, we have no plans to stop there. Is it too soon to start exploring some of the technology futures of VSAN? We don’t think so!
This week we are excited to also announce the intent to start a limited beta later this year to test out possible future Virtual SAN technologies (read below for details on how to apply). Looking ahead to the upcoming VSAN beta, there are two primary themes we are focusing on for the release: space efficiency and advanced data integrity.
We originally approached Virtual SAN with a number of key requirements. It had to deliver high performance, simplicity, scalability and of course high availability – basically all the characteristics associated with enterprise-class, traditional storage and then some. In order to achieve the level of high availability we expected, we mirrored data across nodes, which sacrificed space efficiency in order to achieve the level of availability and reliability customer demand. And we’re happy to say from our greater than 2,000 (and growing) customer base, we’ve had zero reports of any customer losing data from Virtual SAN in production, so we’ve been doing something right!
As we look ahead though, in addition to enterprise-class availability, customers may also need greater space efficiency for their use cases. In our upcoming beta, we plan to test two key features that could increase usable capacity by up to 16 times while still delivering the same levels of availability!
First, we can finally provide a better answer to our customers requesting deduplication in Virtual SAN. With this beta version of VSAN, we expect to test inline deduplication that will be applied on a per-Disk-Group basis so that you can still enjoy the benefits of fine-grained, policy-based management. As with any implementation of dedupe, the impact of dedupe will vary based on the workload, but we expect VSAN deduplication to deliver up to 8x improvement in usable capacity.
Next, we plan to test erasure coding with RAID 5 and RAID 6 options. That means the 2x overhead from simple mirroring would drop to 1.33x when you want to tolerate a single failure (FTT=1). For FTT=2, the overhead would drop from 3x to 1.5x. That means that with no changes to hardware, you could see an increase of 50% or 100% in usable capacity with FTT=1 or FTT=2, respectively.
If space efficiency is your top priority, you could even combine deduplication and erasure coding to achieve anywhere from 1.5x to 16x improvement in usable capacity, a massive increase in efficiency that will further add to the cost advantage of VSAN.
While not as sexy as deduplication and erasure-coding, data integrity is critical and goes back to our commitment to deliver a highly-available and highly-resilient storage solution. To that end, we are excited to announce that we plan to test end-to-end, software checksums in the beta as well. The goal is to protect against storage bit rot, network problems, software and firmware issues. The checksum will use CRC32c, which utilizes special CPU instructions thanks to Intel, for the best performance. These software checksums will complement the hardware-based checksums available today.
The software checksum will be compatible with all VSAN data services, including vsanSparse snapshots/clones, deduplication, and erasure codes.
If you are interested in being considered for the beta (and apologies in advance that we will not be able to accommodate all requests), then you can summit your information at www.vmware.com/go/vsan6beta.