On Thu, Mar 31, 2005 at 10:24:12AM +0200, CheHax wrote:
> As I read your answer, I am quite confused about the fact that Swish-e
> is not the good solution since I have 25 million records. If swish-e
> is not that best solution, what would be the best one then ? Fulltext
> search in mySQL is not an option, and I'm afraid I'll soon have to
> handle even more records than the current 25M ones.
I guess it's a matter of selecting the right tool for the job. Can
you afford to spend hours reindexing when data changes? Will it
scale as needed? Will you need to index multiple languages?
You could look at the swishd cluster system that Dave Seff posted
about lately, but I still suspect you will run into some scalability
issues. Swish-e was originally designed for indexing web sites --
and I don't think that was a website that had 25 million static pages.
There's companies that try to fill that big data niche -- Google
sells products and there's companies like Inktomi. If you want
something free you might look at
And http://www.searchtools.com/ is always a good place to look.
Unsubscribe from or help with the swish-e list:
Help with Swish-e:
Received on Thu Mar 31 07:08:46 2005