Information needed to decide on compression algorithm is:
* Percentage of digests bits differing between two digests. Should be
quite linear to traffic amount / cache size, but I need some input on
what ranges to expect.
* Locality of the differences. My assumtion is that the hashing causes
the differences to have random distribution with close to no locality.
/Henrik
Alex Rousskov wrote:
>
> Hi there,
>
> We have a simple yet efficient algorithm to generate Cache Digests
> diffs/deltas. Before we start implementing this long awaited feature, we
> want to poll for better algorithms! Sorry, no cash prizes. Details at
> http://squid.nlanr.net/Squid/CacheDigest/diffs.html
>
> Thanks,
>
> Alex.
Received on Tue Jul 29 2003 - 13:15:59 MDT
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:12:16 MST