- - - - - - - - - - - - - - - - - - - - - - - - - - - -
Name: luca pellegrini
Subject: questions and suggestions on dpsearch

Hi Maxime,
i've just started to use DataParkSearch and i'm trying to full-index an amount 
of 1000 URL. I have a powerful server, a dual Opteron with 16Gb of memory and 
more that 50GB HDD, mounting a Debian and using dbMode=crc-multi at the moment. 
i have different questions:
1)dbmode=cache doesnt seem to work; the indexer succesfully indexes the 
specified urls but at search-time, i'm always getting 0 result
2)i can see in the documentation that it's possible to "distribute words index 
and web server between different machines". How can this be performed? Is it 
possible both in cache mode and crc-multi mode?
3)do you think that moving stored data to a "memory based filesystem" instead 
of a traditional file system could meaningly increase the performance?
4)using MySql, is it possible to change the engine of the tables from "MyIsam" 
to "Memory"?
5)suppose that i'm going to index a huge amount of web-sites, for example 
1'000'000 web sites. Can you give me some advices about the configuration? 
Should i modify some config parameters such as expire-time, and so on? 
Looking forward to receiving yur answer(s)!!!!
 Thank you for your kindness

Luca Pellegrini
- - - - - - - - - - - - - - - - - - - - - - - - - - - -

Read the full topic here:
http://www.dataparksearch.org/cgi-bin/simpleforum.cgi?fid=02;post=

Reply via email to