Changes and updates to Neptune Analytics

Query REDUCE function

Added support for the OpenCypher REDUCE function, allowing a list of values to be combined using a user-defined expression.

February 26, 2025

Improved memory management in Closeness Centrality algorithm

In the Closeness Centrality algorithm, a memory/performance tradeoff has been added to enable execution on smaller instances.

February 26, 2025

Fixed incorrect results while setting Map properties from a nested map

An issue has been resolved where incorrect results occurred when setting map properties from nested maps in Amazon Neptune (e.g. WITH {struct: {prop: "val"}} as row MATCH (n:label) SET n+=row.struct RETURN n). This fix ensures that map properties are correctly assigned and retrieved, providing accurate and reliable data handling when working with complex nested structures in graph queries.

February 26, 2025

Vertex lookup

Improved vertex lookup performance and memory footprint.

November 15, 2024

Snapshot creation issue

Fixed an issue that prevented statistics from being included in graph snapshots. Graphs created by restoring such a snapshot will not initially have statistics available, and may experience degraded query performance.

November 15, 2024

Query plan cache update

Disabled query plan cache by default for parameterized mutation queries to avoid potential InternalFailureException.

November 15, 2024

Query improvement - Internal failure

Fixed an issue where using COLLECT(DISTINCT()) returned an InternalFailureException in some cases.

November 15, 2024

Query improvement - Internal failure

Fixed InternalFailureException when user uses a value of unsupported type with aggregation functions (ie sum(<string>)).

November 15, 2024

Query improvement - ID match

Correct ~id match behavior. Invalid ~id values like null or non-string types lead to zero match for MATCH (ie. MATCH (n {`~id`: null}) ) or type error thrown for MERGE/CREATE (ie. CREATE (n {`~id`: null})).

November 15, 2024

Query improvement

Fixed an issue where some limit queries only return partial results during execution.

November 15, 2024

Property support over mixed type entities

Support for property/properties over ambiguous/mixed type entities. This avoids the error for queries failing with "Property access/properties over ambiguous type not supported in this release".

November 15, 2024

Performance improvement

Improved performance for queries that use numeric algorithm output for result ordering.

November 15, 2024

Performance improvement

Improved performance for queries which use large static lists or maps. Certain queries related to vector upsert or search algorithms, where a static list of float values is passed as an embedding, have seen significant performance enhancements. A sample vector.upsert query can be seen here.

November 15, 2024

CALL subquery support

Added Support for CALL subquery, allowing execution of operations within a defined scope. A CALL subquery is executed once for each incoming row and the variables returned in a subquery are available to the outer scope of the enclosing query. Variables from outer scope can be imported into a CALL subquery using an importing WITH clause.

November 15, 2024

Algorithm parameter handling fix

Fixed an issue where passing concurrency=0 to algorithms returned an error.

November 15, 2024

Byte handling

Previously, when data was loaded using the Byte data type it was erroneously stored as an unsigned byte, this has been updated to store the data a signed byte. Previously loaded data and queries will not be impacted.

July 29, 2024

Updated float serialization

Float serialization changed to use a lesser number of digits for higher precision. This will change the float values that are returned from the server.

June 20, 2024

UNWIND performance improvement

Improved the UNWIND operations (e.g. transform a list of values within a property into individual vertices or edges) to help prevent out of memory (OOM) situations.

June 20, 2024

Query plan cache update

Fixed an issue in query plan cache when skip or limit is used in an inner WITH clause and parameterized. This fix ensures that parameter values for SKIP and LIMIT are now properly handled, providing accurate results for every execution. If you disabled the query plan cache, you can now remove the query hint QUERY:PLANCACHE "disabled" when submitting a query.

June 20, 2024

Increased efficiency when handling special characters

Fixed an issue with handling of node and relationship labels, node identifiers, and property names containing the “Ā” character (Latin uppercase letter A with Macron, unicode code point U+0100).

June 20, 2024

Datetime formats

Extended support for datetime formats.

June 20, 2024

Condition key - neptune-graph:PublicConnectivity

neptune-graph:PublicConnectivity filters access by the value of the public connectivity parameter provided in the request or its default value, if unspecified. All access to graphs is IAM authenticated.

April 29, 2024

StartImportTask available

Neptune Analytics now allows you to efficiently import large datasets into an already provisioned graph database using the StartImportTask API. This API facilitates the direct loading of data from an Amazon S3 bucket into an empty Neptune Analytics graph. This is designed for loading data into existing empty clusters.

March 30, 2024

Neptune Analytics available in Europe (London) region

Neptune Analytics is now available in the eu-west-2 Europe (London) region.

March 14, 2024

Column delimiter ; is now supported in values

When a column delimiter presents in the values, e.g. for 'two;words', two values {'two','words'} will be inserted. However, if the column delimiter is escaped by a backslash '\', the value will be inserted as a whole with the escape character removed, e.g. 'one\;word' will be inserted as 'one;word'.

February 27, 2024

Graph provisioning time reduced to five minutes or less

Neptune Analytics graphs are now provisioned and ready to be used in five minutes or less.

February 19, 2024

Query improvements - Data Plane SDK

The Neptune Analytics data API provides support for data operations including query execution, query status checking, query cancellation, and graph summarizing via the HTTPS endpoint, the AWS CLI, and the SDK.

February 2, 2024

Initial release

Initial release of Neptune Analytics.

November 29, 2023