Release Summary > New Features in Actian Analytics Engine 8.0
Was this helpful?
New Features in Actian Analytics Engine 8.0
 
Data Lake-Enabled. AI-Ready. High Performance Analytics.
Data Lakehouse and Open Standards
AI-Augmented Analytics
Developer Experience and Tooling
Performance and Resource Management
Platform and Security Foundations
SQL Functions and Developer Features
Data Lake-Enabled. AI-Ready. High Performance Analytics.
This release marks the evolution of Actian Vector to Actian Analytics Engine 8.0 and delivers a complete, high-performance foundation for the modern data stack, bridging the gap between mission-critical analytics and the emerging requirements of AI. It’s modernized core is designed for high-concurrency analytics, ecosystem interoperability, and in-database AI capabilities.
Data Lakehouse and Open Standards
Expanding high-performance computing to distributed data sets.
Open Table Formats
Native support for Apache Iceberg, Delta Lake, and Hudi allows users to query live data directly from the lakehouse without ingestion. See QUERY DATA SOURCE.
Transparent Result Caching
Automatically caches external metadata and data blocks to provide sub-second performance of local tables for repetitive external queries.
Unified Data Source Abstraction
The new CREATE DATA SOURCE command enables registration of external warehouses and CloudFS buckets through a single DDL query with automated type mapping.
AI-Augmented Analytics
Bringing machine learning and vector search directly into the SQL layer.
Native Vector Support
Introduces a native Vector data type and similarity search functions to enable Retrieval-Augmented Generation (RAG) directly within the engine. See Vector Data Type and CREATE VECTOR INDEX.
Standardized AI Connectivity
Includes MCP (Model Context Protocol) Server, providing a secure, industry-standard bridge for modern AI agents and development tools. See MCP Server Documentation.
In-Database ML Inference
Users can now run and manage ONNX models internally, executing ML inference as a component of the analytical query. See CREATE TENSORFLOW/ONNX MODEL.
Generative AI (GenAI) model prompts a language model (LM) endpoint as part of a SQL query that can be used to generate, classify, summarize, and create Retrieval Augmented Generation (RAG) pipelines. See CREATE GENAI MODEL.
ML Workflow Built-ins
New functions for feature preprocessing (scaling, encoding) and model performance metrics (Precision, Recall, F1) reduce the need for external data movement. New tools, iisugetudflib and import_udflib to deploy standard ML data pre-processing UDF. See Predefined UDFs for MLM.
Unified Container Management (Dockman)
A utility that creates a unified interface for managing Docker-based extensions (UDFs, Spark integrations, TensorFlow) across both Windows and Linux. See Containerized Services.
Developer Experience and Tooling
Reducing friction for application engineering and data science workflows.
Vector Python SDK
A dedicated SDK available through standard package managers (PyPI) simplifies application development using the Python Database API Specification v2.0. See ReadMe and TOML.
Data Science Workbench
A browser-based environment featuring Jupyter notebook and VS Code integration for accelerated model management and data exploration. See Data Science Workbench Documentation
Observability with OpenTelemetry
A new OTel-based monitoring stack integrated with Grafana and Prometheus delivers real-time telemetry for engine performance. See Installing Actian Monitor.
Performance and Resource Management
Sustaining industry-leading speed while increasing operational flexibility.
Engine Optimizations
Enhancements to the SQL Optimizer and bitfilter operators deliver up to 2.5 times faster query execution on TPC-DS benchmarks.
Transient Table Support
Introduces RAM-resident tables for ETL and temporary session-based workloads, improving throughput and reducing Cloud storage garbage collection. See CREATE TRANSIENT TABLE.
Runtime Memory Scaling
Dynamically adjust buffer pool and query memory allocations without system downtime to adapt to shifting workloads. See X100 Configuration Parameters.
Binary Data Support in UDFs
BYTE and VARBYTE values are now supported by UDFs. See Scalar User-defined Functions (UDFs).
Boolean Conversion
Native conversion from BYTE/VARBYTE to BOOLEAN is now supported. See BOOLEAN.
Session Control
SET CALLX100VERBOSE | CALLX100SILENT enables or disables printing of CALL X100 output to the terminal monitor session. See SET CALLX100VERBOSE | CALLX100SILENT.
LZ4 Compression
Support for "lz4" or "auto" compression hints for BYTE/VARBYTE columns. See Function-based Encryption, Encryption Functions, Data Compression, and [cbm] Settings.
Time Zone with Timestamp
All Actian Analytics log files include the time zone and the corresponding timestamp.
Cloud Checkpoints
Enhanced data resilience for X100 tables running in Cloud environments. Cloud backup has been enabled for X100 journaled databases. See Setting up a Cloud Checkpoint Location.
Advanced External Table Staging
Specifies multiple SparkSQL statements or full Scala scripts for complex ETL. It is now supported on both Windows and Linux. See CREATE EXTERNAL TABLE.
Scala UDFs
Scala UDFs are executed through the Spark Provider container which can be downloaded using the iisuspark command. It is now supported on both Windows and Linux. See iisuspark Command-- Download Spark Connector Container Image from Repository.
Platform and Security Foundations
Modernizing infrastructure support and hardening the security posture.
TLS 1.3 Security
Encrypted client-server communication is now enabled by default to satisfy Zero Trust security mandates. See Using Secure Communications Encryption with TLS.
Secure Credential Management
New DDL commands allow secrets to be managed securely by name, removing sensitive information from standard SQL text. See Credential Management.
Expanded OS Support
Added support for RHEL 9/10, Ubuntu 24.04, Windows Server 2025, and ARM64 architectures (preview on demand).
Transaction Log File Encryption
Configuration-By-Forms (CBF) utility is able to configure Transaction Log File Encryption. See Transaction Log File Encryption.
SQL Functions and Developer Features
Feature/Function
Description
TRANSLATE
Replaces a sequence of characters in a string with another sequence. See TRANSLATE.
SUBSTRING_INDEX
Allows to search for substrings within a string. Enhanced with delimiter_orientation to control search direction. See SUBSTRING_INDEX.
IS DATE / IS TIME
Predicates to validate if a value can be converted to Date/Time types. See IS DATE/IS ANSIDATE Predicate and IS TIME Predicate.
Database Comments
Ability to add descriptive comments directly at the database level. See COMMENT ON DATABASE.
Jaro_Winkler for X100
Calculates the Jaro_Winkler similarity between two strings c1 and c2 and returns the float value between 0 and 1. See JARO_WINKLER.
LIKE Columnar Patterns
Column names can now be specified as LIKE patterns (for example, WHERE col1 LIKE col2). See LIKE Predicate.
AES_ENCRYPT_BINARY_IV
Application-level encryption supporting VARBYTE values matching existing Ingres functionality. See Encryption Functions.
AES_DECRYPT_BINARY_IV
Application-level decryption supporting VARBYTE values matching existing Ingres functionality. See Encryption Functions.
INSERT ... EXTERNAL CSV
Supports a HEADER flag to include column headers in exports. See INSERT INTO EXTERNAL CSV.
CREATE TABLE ... CLONE
Supports WITH LOCATION=(...) to clone tables to specific storage locations. See WITH OPTIONS for Table Cloning.
HELP MODEL
Displays detailed information about specified ML models. See HELP.
POSITION(c1, c2, i3)
Searches the position i3 and returns the position of the first matching occurrence at or after that offset. See POSITION.
 
Last modified date: 09/11/2026