Amazon Neptune Engine Updates 2019-10-15

Added a Gremlin Explain/Profile feature (see Analyzing Neptune query execution using Gremlin explain).

Added Support for Gremlin script-based sessions to enable executing multiple Gremlin traversals in a single transaction.

Added support for the SPARQL Federated Query extension in Neptune (see SPARQL 1.1 Federated Query and SPARQL federated queries in Neptune using the SERVICE extension).

Added a feature letting you inject your own queryId into a Gremlin or SPARQL query, either through an HTTP URL parameter or through a SPARQL queryId query hint (see Inject a Custom ID Into a Neptune Gremlin or SPARQL Query).

Added a Lab Mode feature to Neptune that can allow you to try out upcoming features which are not yet ready to be used in production.

Added an upcoming Neptune streams feature that reliably logs every change made to your database into a stream that persists for a week. This feature is available only in Lab Mode.

Updated the formal semantics for concurrent transactions (see Transaction Semantics in Neptune). This feature provides industry-standard guarantees around concurrency.

By default, these transaction semantics are enabled. In some scenarios, this feature may change current load behavior and reduce load performance. You can use the DB Cluster neptune_lab_mode parameter to revert to the previous semantics by including ReadWriteConflictDetection=disabled in the parameter value.

Improved the Instance Status API by reporting what version of TinkerPop and what version of SPARQL the engine is using.

Improved Gremlin subgraph operator performance.

Improved the performance of Gremlin response serialization.

Improved the performance in the Gremlin Union step.

Improved the latency of simple SPARQL queries.

Fixed a Gremlin bug where timeout was incorrectly being returned as an internal failure.

Fixed a SPARQL bug in which ORDER BY over a partial set of variables caused an Internal Server Error.