Comments on: Indexing structured data (again)Mostly programming with a few bits of other stuffBy: Phil Dawes Stuff Blog Archive Some ideas for static triple indexing- April 30, 2007 ... wrote a bit about representing structured data in the last post. Heres some ideas for how I plan to index the ...http://www.phildawes.net/blog/2007/04/29/indexing-structured-data-again/#comment-30728 By: carmen- April 30, 2007 are there any stores that don't support named-graphsquadscontexts and instead focus on providing niceties for a single 'shared context' as you put it. my biggest problem has been stuff like - how do you grab all the children of http:mysitecars without either bypassing the RDF lib and doing "select from resources where id like 'http:mysite.comcars%'" since doing the same in SPARQL is just waaaay to slow, and doing a triple like: is pretty insane. don't get me started on countgroup_by etc..http://www.phildawes.net/blog/2007/04/29/indexing-structured-data-again/#comment-30726 By: Nick Johnson- April 29, 2007 I'd consider using an n-ary structure for your indexes instead of just a sorted list: Sort your index, then take every nth item (sized so the number of items returned just fits in a single disk block), and create a node from them. Recurse for each subdivision thus created. This will give you _much_ better locality of reference when you access the index, so for an infrequently used index, you won't have to read nearly as much into memory.http://www.phildawes.net/blog/2007/04/29/indexing-structured-data-again/#comment-30525 |