Hello avid readers 
,
Today I got a very rough proof of concept regenerative distributed hash table (DHT) working. It's a peer-to-peer setup. When the client makes a request, it queries the master server (version two will make it true p2p) and gets an address of where the data corresponding to the hash key is. If that server is down, or the server doesn't respond, the client returns to the master and gets the next host that has a copy of that key slice. If that host is up, the client downloads the entire keyslice, and tells the master server that it's now replacing the down host. If there are enough clients making requests, the downed servers should be replaced fast enough to keep the data intact.
Currently, it will only allow users to query and regenerate, I haven't added any logic to allow for updates or additions, but I'm going to keep plugging away at it. I also want to add some logic into the master server to be able to split up the entire DHT once it's redundant enough, that way each server has reduced strain.
Peace and chow,
ranok