The UDF engine startup is now part of the ingstart utility, which ensures predictable behavior and can surface configuration issues quickly and efficiently. This change enables the Configuration-By-Forms (CBF) utility to control the UDF engine startup and enables the ingstatus utility to view the UDF engine status.

See cbf Command--Start Configuration-By-Forms in the Command Reference guide.

The Spark Connector 3.0 allows you to pre-process data you import (e.g. JDBC) from external tables and Spark. You can apply this for data cleaning, schema mapping, etc., before loading data into Actian X.

This feature is supported only with Actian X Spark Connector 3.0 in conjunction with Spark 3.1.1

See Load Data with Spark SQL through the Spark Connector in the Database Administrator Guide.

This feature adds the ability to automatically rotate the X100 error log file if its size surpasses a specified limit or if a specified time interval has passed. It also adds the option of automatically compressing the rotated log file to .gz format.

See Automatic Log Rotation in X100 in the System Administrator Guide for more information on this feature.

This feature alters the behavior of the x100_partition_scheme “Auto” to use different partitioning schemes. The x100_partition_scheme dbms parameter determines the default partition scheme to be used if the WITH PARTITION clause is not specified when an X100 table is created. See Ad hoc Auto Partitioning in the Database Administrator Guide.

For all new installations of Actian X 12.0, the inbound and outbound encryption mode (ib_encrypt_mode and ob_encrypt_mode) is set to ON by default.

See Configure AES Encryption (Ingres Net and GCA Remote Access) in the Security Guide.

This feature adds the ability to create and store a checkpoint on a cloud provider. Cloud checkpoints are created using the open source rclone utility. See Cloud Checkpoint Locations in the Database Administrator Guide.

This feature provides a helm chart to deploy the Actian X Kubernetes Operator which can then install Actian X into Kubernetes. The ActianxDbms CR has the ability to start in Disaster Recovery (DR) mode. This means a primary and standby container will be created. IngresSync will sync changes in the main database on the primary to the standby.

Containerized Python and Java UDFs are release features that can be used in a production environment. You can choose to install them automatically while installing the product.

This option adds the ability for loading data into and out of X100 tables. The -with_vwload option, in conjunction with -with_csv, transfers data from a source database to a target database using the copydb command.

During execution of a COPY VWLOAD query, the COPY VWLOAD Load Statistics feature shows the current progress status of the loading process or shows if the load has hung up. After the load job completes, the Results Statistics can be queried after executing the COPY VWLOAD query to obtain the load statistics for the particular query.

Specific cryptographic hash functions are now available for use with X100 tables. The newly added functions include MD5_HEX, SHA1_HEX, SHA224_HEX, SHA256_HEX, SHA384_HEX, SHA512_HEX. The specific functions will work only for specific table type. For example, the function “MD5SUM” will not work against X100 tables.

The ELEMENT function splits the input string into elements separated by a specified separator and returns the specified number of elements concatenated by separators starting or ending at an element of a given index.

The ELEMENT_COUNT function splits the source string into elements separated by a specified separator and returns the number of elements within input source string.

This feature enables pattern-matching on more complex patterns by providing the standard conforming SIMILAR TO function that works on SQL standard regular expressions, and SUBSTRING function with pattern-matching predicates such as [LIKE, SIMILAR TO, BEGINNING, ENDING, LIKE_REGEX] functions that works on the X100 engine.

See Pattern-matching Predicates in the SQL Reference Guide.

In a multi-user setting, UDF management can allow a common pool of UDFs created by the Database Administrator (DBA) to be accessible to all users of the database.

This feature enables the ability of the FIRST n qualifier to be used in sub-selects as well as the main query and it is widened to include the offset capability using FIRST n AFTER m as a SELECT qualifier for both X100 and Ingres tables.

Actian will no longer be adding new features or functionality to Ingres Replicator. Actian will continue to support the existing functionality in Ingres Replicator.

See Ingres Replicator in the Replicator User Guide.

This feature adds the ability to synchronize the changes between two or more copies of a database in different installations. See IngresSync Tool in the System Administrator Guide.

QBF now supports the data types IPv4, IPv6 and Universal Unique Identifier (UUID).

The IS BOOLEAN predicate can be used to determine if a value can be converted into a Boolean data type.

The lengthb() function is an alias for the octet_length function and has been provided to ease migration efforts.

Performance Testing is now a part of the Actian X Release Cycle.

A Performance Test Harness has been designed to run multiple benchmark tests to establish performance comparisons for each new release and the prior release of Actian X. These tests include, but are not limited to, TPC-H, TPC-DS and TPC-C workloads.

The Performance Test harness is run against both heap (transactional) and X100 (analytical) table types before release and any significant performance degradations are addressed before release.

Actian X 12.0 testing showed no performance degradations over the previous release, in fact it showed a small improvement in performance.

The performance enhancements for GCC Het-Net Processing include:

The Net Server functionality has been incorporated into the Data Access Server (DAS). DAS can now handle both Ingres/Net and thin client (JDBC/.Net) connections, either on a single port or can be configured to support multiple ports including standard Net and DAS ports.

This feature adds the ability for a user to have the (PROTECTED_USER) privilege. A user with the (PROTECTED_USER) privilege cannot be impersonated by other users, and the user cannot be dropped, altered or have its password changed by other users, except the installation owner.

When a query is executed, the result is persisted (that is cached). The result is purged from the system if there is limitation to the memory allocated, or when the result is invalidated due to an update.

This feature can substantially reduce query time because Actian X bypasses query execution and instead retrieves the result directly from the cache. Query result caching will improve performance only when there are multiple requests of the same select query. Typically, query results are reused if certain conditions are met.

The Spill to Disk feature enables configuring query results to spill to disk when the query result memory is low.

See Query Result Caching in the System Administrator Guide.

This feature adds support for different URL schemes, such as gs, adl, adls, abfs, abfss, s3a, s3n, http, https, hdfs, maprfs, and viewfs, to COPY VWLOAD (SQL load), vwload (command line load in cluster and non-cluster mode) and INSERT INTO EXTERNAL CSV (SQL export) by enabling and deploying Apache Hadoop file system drivers for Actian X.

See Methods for Loading Data in the Database Administrator guide.

The SmartMinmax feature permits the database to actively select a current subset of MinMax index columns. Actian X drops unnecessary index columns and adds necessary columns to the MinMax index according to their predicted selectivity, with respect to upcoming filter queries, based on the set of filter queries seen so far.

This feature reduces memory utilization, while attempting to maintain run time performance, by eliminating index columns that are rarely queried or do not have a high (predicted) selectivity.

See Smart Minmax in the SQL Reference Guide.

X100 table structures now support String Truncation Warn.

Renaming partitioning key columns in the partitioned tables is now supported for X100 tables. When a X100 table is created with "hash" partitions, the columns on which the hash partition is set can be renamed.

Cloning a table is functionally equivalent to creating a full copy of an X100 table or a metadata copy of a Ingres table. Once the clone is created, they are two separate, independent tables. Subsequent modifications or alterations to one do not affect the other. See Table Cloning in the SQL Reference Guide.

Table Synonyms can now be used to load data into the table.

Improved error messaging for Syscall Setconf. This is currently supported for X100 tables only.

This feature adds the ability to support the IngresType “UUID” to the .NET Types.

This feature adds the ability to support the “UUID” data type to the ODBC driver.

Enhancements to the Workload Management feature provide the ability to queue workloads and more finely control how workloads are executed.

For better control over queries, the database administrator can configure workload management queues for users, groups, and roles. Queues can then be configured to allow queries of certain users, such as a CEO, to be executed before other queries. System defined queues exist to allow certain types of queries, like short running queries or queries against system catalogs, to be executed at a higher priority.

The workload management enhancements include:

To set up the workload management system, the DBA can enter values into three new tables in the iidbdb database or use the ALTER WLM statement to create the workload management configurations.

New SQL elements include:

See Workload Management 2.0 in the System Administrator Guide.