Hi folks,

I'm considering spiking hibernate search as a replacement technology for the stored procedure base matching system I'm working on currently. The system does a lot of signed numeric range based matching and I was pleased to see that hibernate search now supports the recent lucene trie based numeric range search.

I'm fishing for some opinions or if possible even some examples so I can do a little demo. I'm hoping that it will out perform the current procedural SQL approach by a large margin but I don't know enough about it to predict this accurately. The database concerned is MS SQL server.

Could we expect matching to be performed faster through Lucene than in stored procedures in the database?

Thanks in advance for any opinions.

Barry.

Hi Barry,

I think it is very hard to give a general answer to this question. I don't think you will get around building an actual prototype.
There are so many parameters which can impact the result, eg the hardware - Running the Lucene index on a crappy machine, while MySQL runs on a high performance server. A lot will also depend on the proper configuration of Hibernate Core and Search. There are a lot of parameters which you can tune to make things faster. And at the same time a wrong configuration can slow you down.

You should be able to build a minimal indexing and searching application with very little code.

--Hardy