macOS port?

[lateworm]

Could there be a macOS version of this at some point? I could see it being really useful to a lot of the developers who are on Macbooks. Even with fusion drives and SSD's, I often hear complaints about lack of space, speed and having to bring external storage because people have big, complex dependency trees that use absolute paths and are very tedious to sync.

[Christopher (Drashna)]

Unfortunately, we don't have any plans on doing something like this right now. 

 

Though, I'm not sure how much this would help, as you'd still need a decent amount of local disk space for the cache.

[lateworm]

That's ok. I haven't tried this on Windows yet and was a bit overexcited when posting. I'm probably unclear on what it actually does. For what I'm after, I had this simple idea of a virtual disk that only caches actual files on access requests, and then has options to avoid the cache accumulating or becoming stale. Like erase synced cache after X time of disuse. Thanks.

[Christopher (Drashna)]

It uses the specified cloud providers to store raw (or encrypted) disk data that's been broken up into chunks. 

 

This means that the data on the CloudDrive disks are only available using our software,.  But it shows up and acts like a normal, physical disk. 

 

Additionally, it does cache the data, but retains it based on the cache settings, and disk usage (it will prune unused data, but retain frequently accessed content).   It does include a prefetching option to optimize access, and pins certain data (such as NTFS data and folder structures) to optimize access. 

 

This part is managed automatically, and not based on any user configurable values.  But you can specify the cache size (and the type of cache). 

 

 

To be honest, if you want to check it out, the software includes a 30 day, fully featured trial period, so you can test it out. 

[lateworm]

Thanks a lot for explaining. Maybe I can make good use of it. For Mac, I'll try to make a DIY https://en.wikipedia.org/wiki/Fusion_Drive where the large/slow part is a 20 TB network volume. At most ~50 GB would have to be cached when in use (totally pruneable before caching some other portion). There's a lot of automagic with https://en.wikipedia.org/wiki/Core_Storage , but provided it actually works, it's ok if I have to trigger pruning manually or something like that. As long as there's some indicator that it needs to be done. I'll try CloudDrive when I'm on Windows and maybe find uses for it there. It'll be a great reference at any rate.