algorithm - Copying files in a distributed system so that all servers have a copy of all files -
full disclaimer: interview question:
there m-machines. need copy data-sets among these m machines each other each server has copy of data-sets. optimal algorithm this?
i know can solve problem in o(mn) (where n average number of data-sets on each machine) iterating through each server. there better approach?
what self-replication system?
http://en.wikipedia.org/wiki/self-replication#a_self-reproducing_computer_program
e.g.; if have m=100 machines, each data set have:
1tic: 1machine data 2tic: 2machines data 3tic: 4machines data 4tic: 8machines data 5tic: 16machines data 6tic: 32machines data 7tic: 64machines data 8tic: 64machines data 9tic: 100+machines data
i think less complex o(mn)