public distributed filesystem 4863

kevincw01

It's an interesting idea, with the fascinating question of how to keep the system secure and ensure privacy. I think it's a solvable problem.

To start with, you want your files to be encrypted, obviously. You'd use some sort of private key encryption to secure your files. I think the most sensible thing would be to cluster files into "file groups" which are encrypted together.

Then, these file groups could be split up into blocks, along with parity blocks for redundancy. Lookup "par2" for the basic concepts. The basic idea is that you have N data blocks and M parity blocks, taking up a total of N+M space. To recover the original data, all that's required is to have a total of N of the blocks--any combination of data and parity blocks is good, so long as the total number of them is at least N. These data blocks are then distributed to peers, ideally a different peer for each block. This minimizes the chances of losing data and also makes it difficult for others to access your data.

To fairly balance load, perhaps the system could be set up to exchange data blocks--you only get to send a data block to a peer if you accept one of his data blocks in return. Periodic random checksum queries could confirm whether a peer has honestly kept your data block in storage.

Isaac Kuo