[EMAIL PROTECTED] wrote: >Hello, > >some more thoughts about parallel content-aggregation. I plan to use >content-aggregation for aggregation of external data, caching has to >be disabled because of frequent changes in the delivered data. Because >the external xml-source is not very fast I expect the performance benefit >of a >parallel mechanism to be more than same milliseconds. > >The ordering of incomming SAX-Events could be handled by the separation >of connect-phase and deliver-phase. During connect-phase (connecting to >foreign host, waiting till first byte arrives) all threads work parallel, >but the >deliver-phase (streaming into handlers) is synchronized by a big lock >shared >between all threads. > >Thoughts ? > >Ciao > Nico Dirks > I have the same situation with DB Prism Generator for Cocoon 2. I think that a parallelel content agregation could be produce a better response time of DB Prism architecture because all request return dynamic data (not cached). Dividing the generation in two phased ( Setup and Generated methods) for example: - If in the first phase Cocoon calls to Setup method of all the request in the agregation parts, this will be permit to DB Prism generator parallelizar all these request. - In the second phase Cocoon could be call to Generate method sequencially. Is it mechanism possible? Best regards, Marcelo.
-- Marcelo F. Ochoa - [EMAIL PROTECTED] Do you Know DB Prism? Look @ http://www.plenix.com/dbprism/ More info? Chapter 21 of the book "Professional XML Databases" (Wrox Press http://www.wrox.com/) Chapter 8 of the book "Oracle & Open Source" (O'Reilly http://www.oreilly.com/catalog/oracleopen/) ----------------------------------------------- Lab. de Sistemas - Fac. de Cs. Exactas - UNICEN Paraje Arroyo Seco - Campus Universitario (7000) Tandil - Bs. AS. - Argentina Te: +54-2293-444430 Fax: +54-2293-444431 --------------------------------------------------------------------- To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, email: [EMAIL PROTECTED]