'Provisioned space' doesn't necessarily equal 'used space', so you should find out what the current state of this is and plan accordingly with regard to what the usage (and expected growth) is if you are going to move from Thick in current environment to Thin in vSAN environment.
Minimum storage and the configuration required can loosely be determined as (VM used space x 2) + slack space (VMware recommendation is 25-30%).
Use our tool to look at this in more detail and to account for other usage such as .vswp:
More information on this:
thanks for your response Bob.
there are mix of thin and thick vms.
for thick vms we can take the provisioned space for vsan storage calculation,
but how do we include thin as iam not sure to what extent they grow??
"but how do we include thin as iam not sure to what extent they grow??"
Well you could err on the side of caution and allocate enough space for their full provisioned space.
You could likely get estimates on growth over time if you have relatively long-term back-ups and compare their size to something more recent and current usage (do note though that it could have been TRIM/UNMAP/punch-zeroed in the interim) and/or talk to the VM owners/users. Another not so straight-forward method would be if you have old support bundles but if you wanted to look at granular per-VM data (and not just df_-h.txt) you would probably need bundles from all the hosts these data set resided on.
I forgot to mention (as you didn't ask ), also account for capacity used by .vswp - these are Thin by default in later versions so they really shouldn't be using much space (provided your hosts have adequate memory and they never need to swap to disk), by default these are 2x the provisioned memory minus reservation.
thnaks Bob.for thin disk iam taking 20 percent extra in total raw storage calculation as this is what i have been informed .
i am going to add other requirement also like swap and mirroring of each vm as its going to be hybrid.
if could comment on following
i think we can have thick disk on vsan cluster though default is thin.
also what is slack space ?
thanks iam checking this .can yu tell me how many host failure cluster can tolerate has been included in this calcultion.
One failure to tolerate
Thanks ,i am going to check this however i thought of using powercli also .
if yu can suggest me the raw space required for folowing
i am using below calculation considering hybrid .
2*(sum of provisioned space)+30 percent of sum of provisioned space+swap space .
can yu tell me what is wrong in above calculation.
Sorry, can you more explain your configuration ?
And if you share the total used space needed.
Per example, the below configuration is for a usable 30TB (with 4 hosts and one to tolerate - hybrid ):
As I understood from your screenshot, you have 5 datastores: 4*27.2TB + 1*44.11GB = around 110 TB. please check the following config:
but this config need 18 disk bays per server (i guess this option is rarely available on HW). so we can increase the size of capacity disks to match the available bays. the below is with 10TB each capacity disk:
you can try this tool with multiple scenarios in order to find your suitable one.
thnaks Hassan iam checking this .
ur most welcome
Please consider marking this answer "CORRECT" or "Helpful" if you think your question have been answered correctly.