AI-driven resilience for IBM i

Predict failure.
Prevent disaster.

World-class IBM i solutions powered by artificial intelligence. High availability, real-time data replication, and flash backups that anticipate threats before they ever cause downtime.

~0sRPO lag
99.99%Uptime SLA
24/7AI monitoring

Built for enterprise stacks

DB2 for iPostgreSQLOracleAWSAzureIBM PowerKafka

The Problem

Your IBM i data infrastructure faces three critical obstacles

Organizations with IBM i at their core need operational continuity and real-time data access. Traditional approaches no longer suffice.

Fragile & costly HA

Solutions based on triggers or application-level replication are fragile during outages, generate locking conflicts, and require constant maintenance. A replication failure can go undetected until it is too late.

Production intrusiveness

Every trigger added to the production system consumes CPU cycles, slows down transactions, and forces changes to applications that have been running for decades. The performance impact grows with volume.

Data trapped in IBM i

Critical business data remains locked inside DB2 for i. Moving it to PostgreSQL, Oracle, or the cloud for analytics, microservices, or gradual migration requires batch processes, nightly ETL, and hours of latency.

The Solution

Three steps. Zero intrusiveness. Continuous replication.

01

Capture

Native OS journaling

The engine connects to IBM i's native journal. No triggers, no application changes, no production impact. The operating system already records every change; we simply read it.

Journal-based CDC via OS/400 journaling
02

Transport

Remote Journaling (RMTJRN)

Changes are transported to the replication server via Remote Journaling, IBM i's native mechanism for distributing journals between partitions. Minimal latency, optimized bandwidth.

RMTJRN with compression and flow control
03

Apply

Go engine, multi-dialect

An engine written in Go applies changes to the target while preserving transactional order, COMMIT/ROLLBACK semantics, and source atomicity. The same engine writes to DB2 for i, PostgreSQL, or Oracle by simply switching the adapter.

Exactly-once, idempotent, with automatic resume

Differentiators

What sets us apart in the IBM i market

Key advantage

Heterogeneous

IBM i DB2 to PostgreSQL, Oracle, or the cloud. The market leader (MIMIX) only supports IBM i to IBM i. This engine breaks that barrier: one engine, multiple targets.

Non-intrusive

Zero changes to production applications. Activated at the operating system level via native journaling. No triggers, no exits, no recompilations.

Banking-grade guarantees

Transactional atomicity, exactly-once delivery, exact resume after crashes, cryptographic self-healing, and hot DDL. Designed for mission-critical environments.

Modern Go engine

Portable, no heavy dependencies, with built-in observability for live lag and RPO monitoring. Static binaries that deploy without a runtime or framework.

Capabilities

Technical guarantees that mission-critical demands

Every capability has been verified with automated test suites, including race condition detection.

Exactly-once delivery

Every change is applied exactly once at the target. No duplicates, no losses, even after an engine restart.

Transactional atomicity

Honors source COMMIT and ROLLBACK. Transactions are applied in full or not at all.

Automatic resume

After a crash, the engine picks up exactly where it left off. No full rescan, no loss window.

Cryptographic self-healing

Per-block cryptographic signatures to detect and correct data corruption in transit or at the target.

Hot DDL

Schema changes (ALTER TABLE, ADD COLUMN) are replicated without stopping replication or losing data.

Live RPO

Real-time monitoring of lag and effective RPO. Dashboards and alerts to meet continuity SLAs.

Object replication

Beyond data, it replicates IBM i objects: data areas, data queues, and system structures.

Use Cases

Scenarios where this engine makes the difference

Banking HA/DR

Banking, insurance, government

High availability and disaster recovery with near-zero RPO and RTO in minutes. Continuous replication between IBM i partitions with automatic switchover and zero transaction loss.

  • Near-zero RPO
  • Switchover in minutes
  • Zero production impact
  • Full transactional audit trail

Heterogeneous migration

Platform modernization

Move data from IBM i DB2 to PostgreSQL or Oracle in real time, without nightly batch or traditional ETL. Ideal for gradual migrations where both platforms coexist for months.

  • IBM i to PostgreSQL live
  • IBM i to Oracle live
  • Coexistence during migration
  • Rollback without data loss

Modernization without downtime

Data & platform teams

Feed data lakes, microservices, or modern APIs with fresh IBM i data. Without stopping legacy applications, without maintenance windows, without batch.

  • Real-time CDC to the cloud
  • Feed data lakes & warehouses
  • APIs on fresh data
  • No maintenance windows

Architecture

End-to-end replication flow

Ports and adapters architecture. The core engine is independent of the source and target SQL dialect.

01

Source

IBM i (DB2 for i)

Production applications write data. The native OS journal records every change.

02

Journal

Native OS/400 journaling

Sequential record of all operations: INSERT, UPDATE, DELETE, DDL. No triggers.

03

Transport

Remote Journaling (RMTJRN)

IBM i's native mechanism for distributing journals between partitions with minimal latency.

04

Engine

Replication Engine (Go)

Decodes the journal, applies transactions with atomicity, exactly-once delivery, and automatic resume.

05

Target

DB2 / PostgreSQL / Oracle

The same engine writes to any relational target by simply switching the dialect adapter.

Comparison

Why not the competition

Traditional IBM i solutions were designed for a homogeneous world. This engine was born for a heterogeneous one.

CapabilityTraditional solutionsOur Engine
IBM i to IBM i replication
IBM i to PostgreSQL replication
IBM i to Oracle replication
IBM i to cloud replication
No triggers / non-intrusive
Exactly-once delivery
Hot DDL
Modern engine (Go)
Live RPO observability

Note: this comparison reflects the architectural capabilities of our engine versus traditional IBM i solutions such as MIMIX and iTera. We do not disparage those tools; they simply occupy a different niche (homogeneous) than the one we address (heterogeneous).

Status & Roadmap

Transparency about where we stand

The project is currently in Proof of Concept stage. The engine is validated; native IBM i integration is being validated. We do not fabricate clients, metrics, or certifications.

Replication engine validated

Completed

Automated test suite with race condition detection. Reproducible demos of every guarantee: exactly-once, atomicity, resume, self-healing.

Native IBM i integration in validation

In progress

Connection with native journaling and Remote Journaling on IBM i. Proof-of-concept tests with real data in controlled environments.

PostgreSQL & Oracle adapters

Dialect adapters for writing to PostgreSQL and Oracle. Interoperability and performance testing.

Observability & dashboards

Control panel with lag, RPO, throughput metrics, and alerts. Integration with existing monitoring tools.

Early adopter program

Early access for selected organizations. Dedicated support and direct feedback with the engineering team.

Contact

Schedule a personalized demo

Our engineering team will show you the engine in action: live replication, failover, and heterogeneous writing to PostgreSQL. No slides, no empty promises.

Technical demo with real data (or yours, if you prefer)

Q&A session with the team that builds the engine

No-commitment evaluation of your use case