yacy_search_server/defaults
orbiter a9cea419ef Integration of the new index data structure IndexCell
This is the start of a testing phase for IndexCell data structure which will replace
the collections and caching strategy. IndexCall creation and maintenance is fast, has
no caching overhead, very low IO load and is the basis for the next data structure,
index segments.

IndexCell files are stored at DATA/<network>/TEXT/RICELL
With this commit still the old data structures are used, until a flag in yacy.conf is set.
To switch to the new data structure, set
useCell = true
in yacy.conf. Then you will have no access any more to TEXT/RICACHE and TEXT/RICOLLECTION

This code is still bleeding-edge development. Please do not use the new data structure for
production now. Future versions may have changed data types, or other storage locations.
The next main release will have a migration feature for old data structures.

git-svn-id: https://svn.berlios.de/svnroot/repos/yacy/trunk@5724 6c8d7289-2bf4-0310-a012-ef5d649a1542
2009-03-17 13:03:27 +00:00
..
autoReCrawl.conf - again an error in defaults/autoReCrawl.conf 2008-08-05 18:57:25 +00:00
performance_dht.profile removed the permanent cache flush and replaced it with a periodic cache flush 2009-01-06 13:51:59 +00:00
yacy.init Integration of the new index data structure IndexCell 2009-03-17 13:03:27 +00:00
yacy.network.freeworld.unit - better memory protection in kelondro caches: computation of needed memory for cache grow 2009-02-15 23:35:59 +00:00
yacy.network.group reduce remote crawl PPM for fresh peers in freeworld to 6 PPM 2008-09-08 09:49:08 +00:00
yacy.network.intranet.unit added more methods to control the vertical DHT (not yet active .. ) 2009-01-23 15:32:27 +00:00
yacy.network.readme added more methods to control the vertical DHT (not yet active .. ) 2009-01-23 15:32:27 +00:00
yacy.network.webportal.unit added more methods to control the vertical DHT (not yet active .. ) 2009-01-23 15:32:27 +00:00
yacy.networks - added another network definition for personal web portals (replacing robinson peers in freeworld) for default use case selection. This solves the problem that the public network freeworld can spoil the personal web index during configuration phase with index entries that the user does not want for his personal web portal 2008-05-23 09:17:30 +00:00