hi,
Sorry for being late on this weekly report. So this week I've work on the complete node splitting ( which was suspend due to slow progress ). I've also worked on optimization on the search onto the DHT, now instead of always restart the search we use cache that we have to search / initialize the search which then speed up the search since we (probably) already know where to search thanks to the cache. For the complete node splitting I'm testing part of it and it seems to work well, it should be merge this week which means that next week I'll ( hopefully ) end the multidimensional key search other the PHT. And then start the documentation of all thoses things for the GSoC. Anyway, see you next week Regards, REYNAUD Nicolas |