[
https://issues.apache.org/jira/browse/JENA-1274?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15809329#comment-15809329
]
A. Soroka commented on JENA-1274:
---------------------------------
The background is the project that originally led me to Jena-- Fedora Commons.
FC is predicated on LDP, so by indirect means, LDP becomes very interesting to
me. A couple of answers to questions:
1) The advantages over dataset-per-resource are minor and major: minor is
compactness, which makes for easier data management, major is ability to fire a
single SPARQL query over the whole panoply of resources.
2) I hadn't thought about union-default-graph: for the moment, I'd like to rule
it out, and come back and think about it more once I'm really sure about the
core mechanics.
3) You're quite right about the name. It is graph-per-tx. (I need to make more
tests to make the behavior very clear.) The WPG name is just baggage. Shall we
call it GPT for now?
4) I actually began with that level of generality but I was having a hard time
find enough time to do it right, so I decided to do something simpler first to
get some problems solved in the smaller space with the time I have. Do you
think it's worth going that far (locking by graphs) and then the kind of thing
that is now here becomes a special case (singleton-only lockset)? Or better to
try to bring this in and see if people use it and then generalize?
5) Right now it plays out slowly. :) The wholemeal dataset stuff ends up being
done as a million tiny transactions.
> Support a writer-per-graph in-memory dataset
> --------------------------------------------
>
> Key: JENA-1274
> URL: https://issues.apache.org/jira/browse/JENA-1274
> Project: Apache Jena
> Issue Type: Improvement
> Components: ARQ, Jena
> Reporter: A. Soroka
> Assignee: A. Soroka
> Priority: Minor
> Labels: ldp, multithreading, named_graphs
>
> Without too much work we could support a writer-per-graph in-memory dataset.
> The target use case here is LDP-style interaction or other RESTful
> architectures, where it is normal for updates to occur centered on one
> resource.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
