Boardrooms kept hearing the same uncomfortable refrain: mission‑critical IBM i applications were stable and irreplaceable, yet digital initiatives demanded cloud speed, customer‑grade experiences, and continuous delivery pipelines that old playbooks could not easily support, creating a high‑stakes gap between reliability and reinvention that no one could afford to mishandle. That tension framed a candid discussion with CloudSAFE leaders Gregg Rohaly and Ron Venzin, who argued the answer was not a wholesale rewrite, but a practical hybrid. Treat IBM i as the transactional core—processing orders, inventory, and financials with Db2 for i—and connect it to hyperscalers for analytics, APIs, AI, and mobile. The result is a pattern already visible in the midmarket: keep the rock‑solid kernel, extend capabilities at the edge, and do both with governance and resilience baked in.
From Core Transactions to Cloud-Scaled Innovation
The working model starts with clear workload placement. IBM i continues to execute core ledgers, point‑of‑sale, and supply chain flows in RPG or COBOL with predictable latency and integrity, while public cloud services shoulder fast‑changing needs. REST or gRPC APIs fronted by NGINX or IBM HTTP Server on Power Systems expose functions to AWS API Gateway, Azure API Management, or Google Cloud Endpoints. Event bridges and queues—IBM MQ, Kafka on Confluent Cloud, or Azure Event Hubs—decouple the two worlds, letting mobile apps, partner portals, or field devices consume updates without touching core batch windows. Data sharing happens through Db2 for i Services, secure FTP/SFTP, or CDC into cloud data warehouses such as Snowflake or BigQuery, which then fuel dashboards, anomaly detection, and marketing automation.
Building on this foundation, development teams layer modern tooling around the core. Node.js, Python, and Java microservices orchestrated in Kubernetes or OpenShift call into IBM i services, invoking business rules already vetted for accuracy. Git‑based workflows, CI/CD runners, and artifact registries run in the cloud while promoting changes to both sides through automated testing gates. This approach avoids risky rewrites and lets teams introduce features––push notifications, digital coupons, dynamic pricing––without destabilizing settlement or fulfillment logic. Rohaly and Venzin emphasized that performance tuning remains non‑negotiable: what crosses the wire must be designed with idempotent calls, caching, and retry semantics, and what stays on IBM i must be profiled for job throughput, I/O bottlenecks, and journaling overhead to keep SLAs intact.
Managed Models: Skills, Economics, and Resilience
Tipping points have gathered speed. Scarcity of seasoned IBM i admins, operators, and developers has made it hard to cover 24×7 operations, OS upgrades, PTF cadence, and security hardening. Capital spend on on‑prem gear also collides with the need for predictable TCO, especially when burst capacity, DR testing, and audit requirements stack up. In response, managed hosting and cloud runbooks now absorb the friction. Providers pool Power capacity, negotiate network and storage at scale, and offer migration plans that sequence cutovers, parallel runs, and rollback paths. FinOps discipline brings showback and chargeback for CPU, memory, and I/O tiers, while policy‑as‑code enforces encryption, MFA, and access least privilege. For leaders under deadline pressure, this eliminates the hunt for every specialist—performance, security, HA/DR, and integration—before moving a single workload.
Resilience reinforced the case. Operators with IBM i DNA are expanding into multi‑platform estates without losing their edge in backup and disaster recovery. Techniques range from VTL with BRMS to SAN‑based replication, PowerHA clusters spanning regions, and orchestration that validates RPO and RTO with evidence—job logs, audit trails, and drill attestation. CloudSAFE’s trajectory mirrors this market shift: partner‑friendly MSPs wire IBM i cores to AWS, Azure, or Google Cloud while co‑managing adjacent stacks for analytics and customer apps. The closing guidance was direct and actionable: baseline current workloads; map APIs around business capabilities; decide data‑sharing patterns before code is written; define RPO/RTO targets with funding and owners; embed FinOps; and pilot one edge service—such as a cloud analytics dashboard or a mobile order tracker—before scaling. Done this way, modernization stayed iterative, budgets held, and the core value of IBM i remained intact.