When Google Drive was first released I thought it could spell the end of the traditional office file server. Simply install Google Drive on each user’s computer, and then the files would be synchronised across each user’s own hard drive. No server required, and multiple backup copies across the office.
But alas, there’s no WAN synchronisation (like Dropbox) and so each file would need to be uploaded by the person who saved it, and then re-downloaded individually to each other user. Lots of wasted bandwidth. Not good.
The other issue is their pricing structure: storage space is allocated per user, not across the whole organisation. The potential issue here is that each user that creates files will be eating up their own GDrive allocation, rather than using a pool available across the organisation.
So: how do we get the advantages of offsite backup and easy access by remote workers, without having to manage large quotas for each individual?
Answer: We keep the server (although it can be a pretty basic one as we’re only sharing files internally) and install GDrive on that. Map any folders inside the Google Drive folder internally as you would if you were running a regular server. Internal users access the files as they always have (by using a drive letter in Windows for example) and the single account that is logged in to GDrive on the server uploads a single copy, regardless of which internal user updated it.
Remote workers can then access these files via their own GDrive installation, and any files created by them will count towards their quota, but at least you won’t need to worry about this for all your users – only your remote users who create new documents.
- One main Google Drive account with an appropriate quota allocation that serves the internal office users
- Always up-to-date offsite backups
- Remote workers can access the latest copy of everything without having to “remote in” to an office server
I have started to convert our office over to this, but as we have close to a terrabyte of data it’s taking a while to do the initial upload. Hopefully once that’s done though, we shouldn’t notice any performance issues during regular day-to-day use.