When setting up Vault Cache you can indicate the size of the cache that you want to allow on end-users workstations. Picking the number is a balance between space consumed on the workstation, space available, amount of impact you want to have on the EV server in order to build the data, and the ongoing load on the EV server keeping the data up to date, and retrieving missing data when a user requests it.
As described previously, the Vault Cache consists of :
a/ A metadata cache
b/ Archived content (cache, I guess)
The size that you set in the policy, or override on a per workstation basis, is the total of these two aspects. There is one caveat though. Your MDC will always be downloaded, full fidelity. What that means is described in the examples below :
If you limit the size to 2 Gb, and your MDC file is 250 Mb, then you can have up to 1750 Mb of archived content.
If you limit the size to 2 Gb, and your MDC file is 3 Gb (yes that’s a BIGGGGG old archive), then you will consume 3 Gb of space on the workstation
Unlike offline vault, vault cache will get the newest data first, meaning that if your entire archive doesn’t fit in to the allotted space, with your metadata cache as well, then at least you’ll have most-recent-data in the content part. The theory being you’re most likely to want to access the most recent data, rather than the stuff from a long time ago.