I'm currently designing an all flash vsan configuration for our environment.
My question is, what is the acceptable deduplication and compression ration being used by VSAN (SImplivity can guarrantee 10:1)?
Hello,
what Simplivity guarrantee is not so trustworthy. (I implement a lot of Simplivity systems in the past) because it depends what kind of datas you have.
If you store JPEGs, MP4 and encrypted containers, than you are far from 10:1.
You have to analyse your datas. In normal environments
Dedupe and Compression ratios as mentioned by marketing departments are always made with perfect conditions. For example 100 vdi images that are exactly the same and never been logged into.
have a read of this PDF it may help you out a bit
https://storagehub.vmware.com/export_to_pdf/vsan-space-efficiency-technologies
ChevUribe,
There are a lot of variables that play into this. Apart from the obvious (common blocks), we need to consider other aspects. Although dedupe and compression is enabled at the cluster level, vSAN does not do global dedupe, as we don't want to have a single point of failure (global dedupe table) that if corrupted, will cause all your data to be lost. With vSAN, the deduplication and compression happens at the Disk Group level. So the number and size of your disk groups will also play a part on these ratios.
If you are using VM encryption, then dedupe/compression will not have an impact for such VMs. But if you use vSAN encryption, dedupe and compression are not affected; and work in conjunction with said feature.
Workloads are also one of the variables that plays into this. For VDI workloads, your dedupe ratio can be 10:1 for example, but a DB may not be a suitable workload for dedupe and compression.
Looking at all these variables during your design phase will help you evaluate pros and cons of DD/C.
As mentioned by others, storagehub.vmware.com has a lot of information that will help you make your decision.
Cheers
Thank you all for your responses!