Re: Exploiting multi-core CPUs during XML parsing
What kind of application would seriously benefit from this sort of fine-grained intra-task parallelism? Server apps typically runs many tasks concurrently, and so embarrasing parallelism is found naturally at more coarse grained levels, which is more cost effective. Plus, given that the algorithm is deviating from pure sequential access it might not even speed things up at all. Wolfgang. On Mar 31, 2006, at 11:22 PM, Sean McGrath wrote: > I have sketched out an algorithm for fast XML WF parsing utilising > two threads that each start at opposite ends of the octet-stream > and meet in the middle. The algorithm hinges on the fact that > start- and end-tags are balanced. i.e. as one thread reads forward > looking for foo start-tag, the other thread is reading backwards > looking for foo end-tag. > > This also has the nice side effect of giving you accurate error > messages quickly. i.e. as soon as a mismatched tag is found, it can > be reported. This is particularly useful with recursive element types. > > Modern machines (even desktop boxes) will increasingly have multi- > core CPUs as standard and I think it behooves us to find ways of > exploiting that in the XML world. > > But to get the most out of the algorithm, you really need a > filesystem that supports reading byte-streams backwards. > > Does anyone know of such a filesystem? > > thanks, > Sean > > > ----------------------------------------------------------------- > The xml-dev list is sponsored by XML.org <http://www.xml.org>, an > initiative of OASIS <http://www.oasis-open.org> > > The list archives are at http://lists.xml.org/archives/xml-dev/ > > To subscribe or unsubscribe from this list use the subscription > manager: <http://www.oasis-open.org/mlmanage/index.php> >
PURCHASE STYLUS STUDIO ONLINE TODAY!
Purchasing Stylus Studio from our online shop is Easy, Secure and Value Priced!
Download The World's Best XML IDE!
Accelerate XML development with our award-winning XML IDE - Download a free trial today!
Subscribe in XML format