2025 Releases

This page documents all Tacnode releases for 2025, organized chronologically with the latest releases first.

2025.5 Engine Release V1.2

Release Date: May 2025

This major engine release focuses on performance optimization, enhanced SQL capabilities, and improved data lake integration.

🚀 SQL Engine Enhancements

Runtime Filter Optimization

• Query optimizer support for efficient runtime filters: Dynamically generates and applies filter conditions during query execution
• Performance benefits: Significantly reduces data scan volumes, I/O operations, and network overheads
• Best use case: Particularly effective for large table join small table scenarios

Stability & Scalability

• Improved SQL engine spill capability: Supports adaptive spill to enhance stability for large workloads
• Enhanced partitioned table usability:
  • Support for truncating parent tables
  • Setting different storage formats per partition (hot/cold tiering)
  • CDC subscription at the partitioned parent level
  • Learn more: Partitioned Tables

Query Optimization

• Enhanced pg_hint_plan support: Runtime query behavior can be dynamically adjusted via the hint_table configuration
• Learn more: SQL Hints

📊 Incremental Materialized View Enhancements

Performance Improvements

• Improved refresh process: Reduced need for table-level locks and better parallel refresh efficiency

New Aggregate Functions

• string_agg(DISTINCT) for distinct string aggregation
• approx_count_distinct for approximate distinct counting
• approx_percentile for approximate percentile calculations
• Learn more: Approximate Computing

Dynamic Configuration

• Storage format modification: Support for dynamic modification of materialized view storage format

  ALTER MATERIALIZED VIEW mv_name SET ACCESS METHOD columnar/row/hybrid;

• Dynamic index creation: Full support for runtime index management
• Learn more: Materialized Views

🔍 Multi-Modal Search Enhancements

Vector Search Optimization

• HNSW index improvements: Now supports vector quantization including int8/fp16 types
• Memory efficiency: Converts original float vectors into more compact types to reduce memory footprint
• Performance gains: Significant improvement in vector search performance
• Learn more: Vector Search

Feature Store Optimization

• Extended window functions support in incremental materialized views:
  • Ranking functions: ROW_NUMBER, RANK, DENSE_RANK, NTILE
  • Window aggregations: SUM OVER, AVG OVER, COUNT OVER, MIN OVER, MAX OVER
  • Distribution functions: PERCENT_RANK, CUME_DIST
  • Position functions: FIRST_VALUE, LAST_VALUE
  • Lag/Lead functions: LAG, LEAD
  • Statistical functions: stddev_pop, stddev_samp, var_pop, var_samp

Geospatial Capabilities

• PostGIS extension: Added support for geospatial queries in SQL
• Capabilities: Spatial relationships, measurement, and geometry operations
• Learn more: PostGIS Geospatial Query

Search Improvements

• Full-text search: Optimized relevance ranking for full-text search scenarios
• Learn more: Full Text Search
• JSON support enhancements:
  • More JSON PATH query push-down optimizations
  • Index acceleration for CAST filter scenarios
  • Partial indexes for sparse JSON fields

🗄️ Data Lake Enhancements

New Table Format Support

• Iceberg Tables: Full support with Iceberg REST API Catalog integration
• Learn more: Iceberg Foreign Table
• Delta Lake: Complete support with Databricks Unity Catalog integration
• Learn more: Unity Catalog Foreign Table

Enhanced Data Type Support

• Array types: Parquet and ORC now support Array types
• Learn more: External Table Type Mapping

---

2025.4 Platform Release: Cache & Auto Suspend

Release Date: April 2025

This platform release introduces intelligent caching and resource management features.

💾 Cache System

• Cold storage acceleration: Dramatically improves access speed to cold storage and data lake
• Performance benefits: Reduces latency for frequently accessed data
• Learn more: Cache

⚡ Nodegroup Auto Suspend

• Intelligent resource management: Automatically suspends Nodegroup instances that are idle for extended periods
• Cost optimization: Significantly reduces compute resource costs
• Seamless operation: Automatic resume when activity is detected

---

2025.3 Platform Release: Catalog & Isolation

Release Date: March 2025

This release focuses on enterprise-grade isolation and multi-tenancy capabilities.

🏢 Enhanced Nodegroup Isolation

• Shared storage architecture: Implements flexible isolation using shared storage
• Isolation types:
  • Read/Write isolation: Complete separation of read and write operations
  • Write/Write isolation: Multiple independent write workloads
  • Read/Read isolation: Isolated read-only access patterns
• Multi-team support: Perfect for multiple teams sharing a database while maintaining operational independence
• Learn more: Nodegroup Isolation

---

2025.1 Engine Release V1.1

Release Date: January 2025

The first major engine release of 2025, introducing groundbreaking features for real-time analytics and advanced data management.

📈 Incremental Materialized Views

Real-time Analytics Revolution

• Near-real-time updates: Incremental refresh provides substantially fresher data with lower resource cost
• Comprehensive aggregation support:
  • Basic aggregations: COUNT, SUM, MIN, MAX
  • Advanced functions: COUNT DISTINCT, variance, standard deviation
• Advanced SQL features:
  • HAVING expressions support
  • WITH clauses (CTEs)
  • All join types: INNER, LEFT OUTER, RIGHT OUTER, FULL OUTER
• Learn more: Materialized Views

🗂️ Advanced Table Management

Partitioned Table Enhancements

• Flexible partition management: ATTACH/DETACH operations for partitions
• Direct data import: Import data directly to parent tables
• Improved maintenance: Streamlined partition operations
• Learn more: Partitioned Tables

Hot/Cold Tiered Storage

• Cost optimization: Balance performance and cost by automatically moving data between tiers
• Reduced storage expenses: Significant cost savings for large datasets
• Intelligent tiering: Automatic data movement based on access patterns
• Learn more: Tiered Storage

🔒 Security Enhancements

Column-level Permissions

• Fine-grained access control: Control SELECT and UPDATE access at the column level
• Enhanced security: Protect sensitive data with precision
• Compliance ready: Meet strict data governance requirements
• Learn more: Column-level Security

🔍 Advanced Search & Analytics

Vector Search Improvements

• HNSW vector index: Dramatically improved recall accuracy and performance
• Billion-scale support: Handle massive vector datasets efficiently
• Enhanced performance: Optimized for large-scale vector operations
• Learn more: Vector Search

Query Optimization

• SQL Hints mechanism: Full support for pg_hint_plan extension
• Query tuning: Fine-tune query execution plans
• Learn more: SQL Hints

🔧 Data Type & Integration Improvements

Enhanced Array Support

• Multidimensional arrays: Full support for complex array structures
• Improved functionality: Better array operations and indexing
• Learn more: Arrays

Integration & Compatibility

• Enhanced tool support: Improved integration with:
  • ByteBase: Database schema management
  • AirByte: Data integration platform
  • dlt: Data loading tool
  • dbt: Data transformation tool

Performance & Stability

• JSON query optimization: Better column pruning and dictionary encoding
• Memory usage optimization: Greater efficiency and stability
• OOM risk reduction: Reduced risk of out-of-memory errors