Try now
Try now
Use Cases
Replication and data migration Zero downtime migration Logical standby Zero downtime upgrade Distributed clusters ''Real Time'' reporting systems and Business Intelligence
Support
Current Feature Set Documentation Professional Services Roadmap
Company
About Us Contact Us Become a Partner News Updates
Try now
Try now
Home
Roadmap

Repstance Strategic Product Roadmap (H2 2026)

  • SQL Server AlwaysOn Read-Only Scale-Out
    • Capability: Allow the Repstance log-mining engine to point directly to secondary read-only nodes or availability group replicas instead of hitting the primary operational instance.
    • Impact: Eliminates the final remaining CPU overhead on primary transactional nodes. High-volume read operations are entirely isolated on the replica.
  • Azure SQL Database & Managed Instance (PaaS) Support and Google Cloud SQL Instance Support
    • Capability: Deploy production-certified native target and source connectors for Azure SQL Database PaaS and Google Cloud SQL (MySQL, PostgreSQL, and SQL Server instances).
    • Impact: Completes Repstance’s cross-cloud coverage matrix, giving the enterprise full multi-cloud and hybrid migration flexibility across AWS, Azure, and GCP.
  • Advanced Bulk CDC Engine (Multi-Statement Grouping)
    • Capability: Redesign the transaction applier loop to intelligently bundle multiple incoming changes (INSERT, UPDATE, DELETE, MERGE) into a single, highly optimized memory batch.
    • Impact: Drastically solves the unindexed column bottleneck. Instead of forcing the target engine to run thousands of sequential Full Table Scans for individual rows, it applies multi-statement blocks as a unified set-based bulk transaction.
  • Intelligent Chunked Initial Load
    • Capability: Split massive history-loading tasks into micro-chunks or parallel data ranges automatically. This feature tracks chunk states independently, allowing for automatic resume if network drift occurs.
    • Impact: Drops initialization times for multi-terabyte data tables by up to 65% while protecting network streams against failure timeouts.
  • Real-Time Data Streams AI Integration
    • Capability: Launch native target output formats engineered explicitly for real-time Retrieval-Augmented Generation (RAG) caches, Kafka event brokers, and AI Vector Databases (e.g., Pinecone, PGVector, Milvus).
    • Impact: Positions Repstance as the essential real-time infrastructure layer for corporate AI applications, ensuring Large Language Models (LLMs) are constantly querying perfectly fresh production data.