Data from push/pull tests, ignoring failures:
AUGUST: 85 seconds (12KB/sec)
SEPTEMBER: 79 seconds (13KB/sec)
OCTOBER: 59 seconds (18KB/sec)

Before August, data is not comparable as we changed the bootstrapping 
thresholds. This is from inserting a 1MB CHK of random uncompressible data to 
one newly bootstrapped test node and then pulling it from another one. Note 
that 10 seconds or so will be waiting for the first block, since we cannot 
parallelise until after that has been fetched, so I consider these figures to 
be rather impressive...

Success rates are pretty good afaik. However, this is data which has been 
pushed and is then immediately pulled from another node - using Freenet as a 
short-term data transfer mechanism, not as a means of storing data. I have 
tests from 1,3,7,15,31 days also, but these have failed to run or insert on 
various days for various reasons, so I need to make a script to check the 
results against whether data was inserted; I will post data soon, preliminary 
results suggest it's a bit better than it has been.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 835 bytes
Desc: This is a digitally signed message part.
URL: 
<https://emu.freenetproject.org/pipermail/devl/attachments/20091022/a537f4c0/attachment.pgp>

Reply via email to