Hi Eliot,
Are all the 10,000 doing seaches all day ? Can you estimate
the required peak number searches per second ?
What about hardware, is a multi server solution practical ? What
kind of hardware do you have in mind ?
What is the expected total size of your data ? How often does it
changed or need to be reindexed ?
Tal
> -----Original Message-----
> From: [EMAIL PROTECTED]
> [mailto:[EMAIL PROTECTED]]On Behalf Of W. Eliot
> Kimber
> Sent: Monday, June 11, 2001 8:40 AM
> To: [EMAIL PROTECTED]
> Subject: [Lucene-users] Performance/Scalability Benchmarks for Lucene
>
>
> All,
>
> We have integrated Lucene into a larger content management system. We
> need to be able to quantify the performance and scalability of Lucene so
> that we can compare it to some commercial systems that we also have
> access to (Convera, in particular).
>
> I searched the archives and didn't find any references to existing
> benchmark tests or results for Lucene. Are there any? We are looking at
> peak scales on the order of 100,000 or 1,000,000 separate indexed files
> being accessed by 10,000 concurrent users (a system with 100,000
> registered users of whom 10% are active at any given time). Users expect
> 5-10 second response times for a simple query (e.g., all docs containing
> the work "spam"). I don't personally know if Convera, for example, can
> meet these scales, but there is some expectation within DataChannel that
> it can (it is currently used with the DataChannel Server portal
> product). We need to know if Lucene can be made to work at these scales
> or, if not, what it's upper scale limits are and/or what would need to
> be done to it to provide the scalability characteristics we're looking
> for.
>
> Of course, we have other lower-scale use cases where we have no doubt
> that Lucene will perform very well.
>
> If you're curious: we've integrated Lucene into a generic CORBA-based
> full-text framework that allows our Python-based versioning content
> management system to use any full-text indexer integrated through the
> framework. We chose Lucene as the first integration because it is open
> source, came well recommended, and appeared to be (and was in fact) the
> quickest way for us to get indexing functionality implemented.
>
> Thanks,
>
> Eliot
>
> --
> . . . . . . . . . . . . . . . . . . . . . . . .
>
> W. Eliot Kimber | Lead Brain
>
> 1016 La Posada Dr. | Suite 240 | Austin TX 78752
> T 512.656.4139 | F 512.419.1860 | [EMAIL PROTECTED]
>
> w w w . d a t a c h a n n e l . c o m
>
> _______________________________________________
> Lucene-users mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/lucene-users
>
_______________________________________________
Lucene-users mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/lucene-users
