The bug fixed in 6312 (the last few builds have been mainly on stats): We were reporting bytes per millisecond and calling it bytes per second on the node status pages! This is why we got the figures of a few bytes per second min and max... http://127.0.0.1:8888/servlet/nodestatus/diagnostics/successTransferRate/{hour,minute,day,month} is a little known diagnostic that gives the average transfer rate for successful files.
Here are the tTransferRate's from my top 5 nodes in the order shown in the nodestatus page: tTransferRate Minimum: 6,101 bytes/second Maximum: 14,361 bytes/second tTransferRate Minimum: 13,178 bytes/second Maximum: 54,665 bytes/second tTransferRate Minimum: 11,894 bytes/second Maximum: 57,544 bytes/second tTransferRate Minimum: 3,778 bytes/second Maximum: 13,809 bytes/second tTransferRate Minimum: 13,522 bytes/second Maximum: 58,103 bytes/second Also, the order of the last estimate's is almost exactly the same as the order they are listed in (which is based primarily on # successful transfers). Nodes without connections, and a couple of brand-new nodes with low estimates, are the exceptions. This also suggests that NGRouting is doing a reasonable job at the node level. Of course network level diagnostics (insert and retrieve latency) are the only true measurement, one that is approximated by FIW/Fishtools insert latency, which I am informed is still pretty bad. That may partly be due to a bad batch of unstable builds which were probably run on many of the network's best nodes, but I suspect it is also to do with the overall level of QueryRejected's. -- Matthew J Toseland - [EMAIL PROTECTED] Freenet Project Official Codemonkey - http://freenetproject.org/ ICTHUS - Nothing is impossible. Our Boss says so.
signature.asc
Description: Digital signature
_______________________________________________ Devl mailing list [EMAIL PROTECTED] http://dodo.freenetproject.org/cgi-bin/mailman/listinfo/devl
