Enterprise file sharing – it’s hard to even write those three little words without wincing a little. As a ragged war veteran of SharePoint administration and various cloud storage product squashing (namely blocking them to stay within legal compliance), I know how difficult it is to execute on a file sharing strategy. And while new players have emerged to help identify and fill the gaps, one was off my radar – until now.
Oxygen Cloud, one of the presenters at Storage Field Day 4, has developed an offering named oDrive that works to enhance your existing file sharing architecture and give it long, cloud like arms that can reach out to endpoints. Rather than syncing data to a cloud device – which has its own barrel of issues – oDrive uses an on-premises gateway to control the sharing of data from your existing storage array to specific users.
The gateway uses REST to serve files between the storage array and agents in the endpoint, using “accelerated file sync” to ensure users have their data. Because the gateway sits between the users and the files, it has the ability to understand when files change regardless of the amount of data that the storage array contains – this eliminates a need to walk/crawl through stale files.
File Syncing as Primary Access
When a folder is shared with someone, they receive an email alert and oDrive notification. They can then choose to accept the invite, at which point the folder structure appears containing pointer files for the items contained within. oDrive triggers a synchronization event when either a user goes to access a file or when the user explicitly desires a folder structure to be synced.
I’m a heavy user of Dropbox, and one of the more annoying parts about it is the fact that it tries to fully sync all files the moment an endpoint is connected. While it’s true that I can disable certain folders from being synced in Dropbox, I ultimately end up losing visibility to those folders. It’s all or nothing level of sync. With oDrive, I can still see the folder and file hierarchy but without the need to fully sync all that data onto my endpoint – only the metadata is sent over initially.
But what about after I’m done working on a set of files? I have years worth of data that live in my file shares. The vast majority is active for a few months and then sits idle (just in case I need it ever again). Rather than waste space on my desktop, oDrive allows you to remove the data and return to simply tracking the metadata for a file or folder. I can see advantages in this – work on a project for a few months, then send it away. If I need it again, the file or folder can be re-synced.
The oDrive system allows for the endpoint to have granular control over what files are kept in sync, and which are kept as metadata. This frees up a lot of wasted capacity on the endpoint. It also enhances security – stealing a metadata file is relatively worthless (unless your file name is “My password is 12345”), while stealing an intact file can be risky. There’s a chance that the thief could decrypt a fully synced file.
There’s also a public perception around theft, especially if you are a customer of whomever had data stolen. People feel more secure knowing that a laptop went missing but little-to-no data was on it, rather than hearing about what levels of encryption were on the device. I can imagine setting policies on the data that is synced to have some files that never permanently live on the device – perhaps for HIPPA or PCI compliance – and are always freshly synced on access. Or other files that have an expiration date to help automatically keep a device clear of older data.
It would appear that oDrive is heavily focused on creating a file sharing experience that is simple and intuitive to the user. This is typically the recipe for success – complex methods with great potential fall to the wayside. I would like to see greater control over who can participate in an environment’s oDrive develop as the solution rolls towards General Availability – such as accounts for consultants and other contractors – but that is on the roadmap.
Overall, I like how much power the endpoint has over the files that are synced and how the folders are presented. Rather than having to deal with a file server named “CHI_WFSP_07X” the user can just name it “Chicago Files” or “Chicago Production Files” – whatever makes sense to him or her. The user can also look at a folder structure and only choose to sync the “Sales” files or specifically a project called “Q4 2013” within Sales.
And, I have to give Oxygen Cloud this – they have a great sense of humor when it comes to buzzwords. 🙂
If you’d like to sign up for the public beta of oDrive – do so here.