[MarkLogic Dev General] clean bufferpool and attribute range index

[Michael Blakeley]

MarkLogic Server doesn't have a buffer pool, nor does it have tables.

MarkLogic Server does have various caches, which are maintained 
independently by each host. You can clear these caches by restarting the 
host (admin > hostname > status > restart). You can learn more about 
these caches from the Query Performance and Tuning Guide 
(http://developer.marklogic.com/pubs/4.1/books/performance.pdf).

Your question about range indexes is addressed by the Admin Guide 
(http://developer.marklogic.com/pubs/4.1/books/admin.pdf). Look for the 
section titled "Using Range Indexes for Element and Attribute Value 
Lexicons". You may also want to read the section of the above-linked 
Performance Guide titled "Optimizing Order By Expressions With Range 
Indexes".

-- Mike

On 2010-04-19 10:23, Ling Ling wrote:
> Hello,
>
> I am running some queries. Among each query, I want a clean buffer pool.
> I don't want the table content or index in buffer pool. How can I deal
> with that? I'm running queries through XDBC in Java. The second question
> is that I have thousands of xml files in database. They have the same
> schema. The document node has a customer node with an attribute id,
> which is an integer. I followed Administrator's Guide 21.7 to set an
> attribute range index to quickly locate an xml document, but it seems it
> doesn't work. The time is still too long. Is there anything I missed?
> Thank you very much!
>
> Thanks,
> Ling
> _______________________________________________
> General mailing list
> General at developer.marklogic.com
> http://xqzone.com/mailman/listinfo/general