The RemoteSearchable class (in the latest CVS) will let you do this. It uses Java RMI to let you search indexes on other machines. With a MultiSearcher you can then search a number of independently maintained indexes on different machines. MultiSearcher searches indexes serially, but it would be fairly simple to extend it to be able to search in parallel, with each search in a separate thread. That should, in theory, scale fairly well, but I don't know if anyone has tried this. If you do, please send a message to describing your experience.

Doug

Piotr Martyniak wrote:
Hi,

It's kind of newbie question - i'm working on a small system allowing to
allocate independent tasks on diffrent computers in local network

I've been wondering if it is possible to use Lucene with system of this
kind - I mean if 2 client-computers get 2 diffrent URLs, is it possible for
them to do indexing job independently and then just send the results to the
server which will connect those with other results.
If it's possible at all, do you think it makes sense to do it on 10, 100,
1000 client-computers? What kind of advantages and disadvantages you see in
solution like this?

Thanks in advance for all your opinions.
Peter



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to