Suggestions for further development

For a more up-to-date list, have a look at the issues.

• horizontal scaling based on Akka(.io) or more likely with the help of Apache Ignite
• provide checkpointing and reapply changes after a power-failure... (full ACID compliance at least on a single computer/node)
• provide a B+-tree for large-index-structures (most probably based on the indirect page-tree), thus a faster COW variant but might be a bit slower for reading (because of leaf-page versioning and indirect pointer access). Version only leaf-pages.
• make index-structures dependent on self-tuning aspects (that is probably test-data sets from which to learn)
• include the index-structures and rewriting rules in the XQuery/XQuery Update Facility processor Brackit
• implement fine grained user-management
• implement n-writing transactions concurrently
• implement mapping from DeweyIDs to unique NodeKeys (for instance to facilitate fast ancestor lookup which is necessary for instance to implement the above item (concurrent write-transactions))
• implement nodes for XQuery which can be reconstructed in memory with the help of the path summary