Github user dick-twocows commented on the issue:
https://github.com/apache/jena/pull/233
I'll try to keep this brief. Two main parts to the PR, Mosaic and Thrift.
Mosaic provides a way to aggregate multiple dataset graphs as one using
optimistic transactional. All threads will acquire a read or write, with
transactions beginning only when a child is touched. This allows indirect
multiple writes and the more parallel write you need just load more children.
Thrift provides access to a dataset graph running in another JVM with full
transaction support. Simplistically âthe Thrift service mimics the dataset
graph interface with the addition of an ID to most calls. The ID allows the
Thrift server to route the Thrift thread pool threads to the correct thread
proxy to maintain thread affinity for remote transaction purposes. The iterator
cache classes specifically the paged based ones cache the quads or triples to
reduce round trips over the network.
If you add Thrift dataset graphs to the Mosaic you get a clustered dataset
graph, i.e. multiple TDBs each running in a separate JVM which can be accessed
using the standard Jena API.
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
