Gluesync 2.2.4: high-efficiency IBM i journal reading and advanced connection pooling

Following the recent focus on multi-pipeline optimization, MOLO17 is excited to announce the release of Gluesync 2.2.4.0. This update follows the significant refinements we introduced in versions 2.2.3.1 and 2.2.3.2.

This latest release cycle marks a major milestone for IBM i Series integration. It introduces stricter resource safeguards to ensure Core Hub stability under heavy enterprise workloads.

IBM i Series: a new high-performance journal reading engine

The headline feature of version 2.2.4.0 is a complete architectural shift for the IBM i agent. We have transitioned from the “classic” Display Journal command to a new approach leveraging native IBM i APIs.

This change allows for significantly more efficient journal reading with reduced system overhead, providing a smoother CDC (Change Data Capture) experience for legacy systems.

Note: Due to this transition, we recommend reviewing the updated permission requirements for the new APIs on the IBM i agent documentation page.

Universal connection pooling and resource safeguards

Building on the connection reuse features introduced earlier this month, versions 2.2.3.1 and 2.2.3.2 have extended our database pool technology to all supported database types.

We implemented a new safeguard to prevent accidental resource depletion. Specifically, it caps the maximum connection count at 25 connections per agent, down from the previous limit of 100.

Administrators can now also define a minimum connection pool size (defaulting to 5) to fine-tune responsiveness versus resource consumption across the entire ecosystem.

Data lake enhancements: custom paths and identifiers

For teams pushing real-time data to Data Lakes (S3, Azure Blob, GCS), version 2.2.4.0 introduces two highly requested features:

• Custom root paths (GSSD-779): you can now specify exactly where data should be stored within your bucket or container. This helps with better organizational structure.
• CSV unique identifiers (GSSD-780): we added a new technical field to CSV outputs to act as a unique row identifier, simplifying downstream data deduplication and processing.

Operational visibility, Core Hub tuning and UI refinements

We continue to refine how Gluesync reports its activity and manages its internal memory.

We also enhanced monitoring capabilities by including a snapshot duration report in notifications (GSSD-590). This field shows exactly how long the snapshot phase took to complete, enabling more accurate historical performance tracking.

The engine under the hood has been optimized for modern container environments.

We increased the default Java Max RAM Percentage for Core Hub from 50% to 75%. As a result, the system leverages more available memory while staying fully configurable via environment variables.

On the interface side, the pipeline management screen has been redesigned to provide more horizontal room for critical operational metrics by optimizing the space allocated to pipeline names.

Reliability fixes

This release cycle addresses several edge-case scenarios to ensure 24/7 uptime:

• I/O performance handling (GSSD-778): Fixed a write-lock issue in the internal agent cache that could occur on disks with insufficient I/O performance.
• Agent ecosystem refinement: we applied comprehensive fixes and enhancements across the board, including MS SQL Server (CDC & CT), PostgreSQL, Oracle (LogMiner & Triggers), MySQL, MariaDB, and SingleStore.

Ready to upgrade?

These latest updates reinforce Gluesync 2.2.4.0 as a robust, enterprise-ready platform capable of handling the most demanding data integration challenges. Whether you are synchronizing legacy IBM i systems or managing massive Data Lake workloads, these refinements ensure your data stays moving with maximum precision and minimum overhead.

Explore the full technical details in the official release notes or book a discovery call to see how Gluesync can accelerate your data modernization journey.