Ivan Shcheklein | 2 Dec 2009 10:51
Picon
Gravatar

Re: Improving Sedna write performance

Hi Alexander,

Thank you for your information.

1. It's better to use attributes for simple data: <Batch id='1'>...</Batch>.

2. Create unique ID for Step  and use indexes to determine insertion point:

<Step id='3.0.1.12'> .... </Step> (which means Simulation 3, Batch 0, Run 1, Step 12)

CREATE INDEX "step" ON fn:doc("whatever')/SimulationData/Simulation/Batch/Step BY <at> id AS xs:string

So, your update will look like this:

UPDATE INSERT ( something ) INTO index-scan('step', '3.0.1.12', 'EQ')

3. Send me just a couple of event logs with your updates.

Ivan Shcheklein,
Sedna Team


------------------------------------------------------------------------------
Join us December 9, 2009 for the Red Hat Virtual Experience,
a free event focused on virtualization and cloud computing. 
Attend in-depth sessions from your desk. Your couch. Anywhere.
http://p.sf.net/sfu/redhat-sfdev2dev
_______________________________________________
Sedna-discussion mailing list
Sedna-discussion@...
https://lists.sourceforge.net/lists/listinfo/sedna-discussion

Gmane