在大规模分布式系统中记录文件

I do a lot of work in the grid and HPC space and one of the biggest challenges we have with a system distributed across hundreds (or in some case thousands) of servers is analysing the log files. I do a lot of work in the grid and HPC space an