How to renew a financial cooperative's infrastructure without disrupting operations

By the Redsis Infrastructure team · October 2026 · 6 min read

A sector that is growing and going digital

Colombia's cooperative financial system closed 2024 with COP 35.67 trillion in assets, up 6.6% year over year, and more than 4.4 million members across 179 institutions, 172 of them savings and credit cooperatives, according to Fecolfin figures reported by La República. Deposits grew 8.7% and surpluses 43%.

That growth comes with more channels: virtual branches, mobile apps, payment network integrations and new credit products. All of it means more load on infrastructure and less tolerance for outages. A cooperative supervised by Colombia's Superintendence of the Solidarity Economy can't afford to have its core or its databases depend on a single server.

The starting point: two platforms at the limit at once

When we started, both of the cooperative's platforms had been in service for about five years and showed familiar symptoms:

• A VMware cluster at 80% utilization, with little room for new projects or organic growth.
• A database on a single server , with no high-availability strategy.
• A core migration that needed more IBM Power processing capacity to go live.
• Long backup windows and backup procedures that needed strengthening on the Intel side.

And one non-negotiable condition: moving to the new infrastructure could not affect critical day-to-day processes.

What we did: renew both platforms as one project

The usual temptation is to tackle each platform separately. Here we designed them together, with a common SAN and storage from the same family, so the cooperative's IT team would gain simplicity, not just capacity.

• IBM Power9 connected to IBM FlashSystem 7200 , with IBM BRMS backup to an IBM TS4300 tape library.
• A three-node VMware cluster on Lenovo ThinkSystem SR630 servers on the same IBM FlashSystem 7200.
• Veeam Enterprise Plus for Intel backup, with a Lenovo server, IBM FlashSystem 5010 storage and a TS4300 tape library.
• A Lenovo ThinkSystem SR650 server with VMware Enterprise Plus and vRealize Operations to provide high availability for the database node.

Five lessons for renewing without disrupting operations

1. Unify storage before you multiply consoles

IBM FlashSystem is IBM's all-flash storage family, with virtualization, replication and centralized management. Using it across both environments lets the cooperative manage all of its storage from the same console. For a mid-sized IT team, that is worth as much as the performance.

2. Review licensing before hardware

One of the project's most financially significant decisions wasn't technical; it was contractual. Instead of buying a new server, the existing IBM Power8 was upgraded to Power9 while keeping the same serial number. That avoided purchasing a GUI software license the current core version used but the new version no longer required. On platforms where licensing is tied to the machine, this kind of analysis can change a project's total cost entirely.

3. Reuse what still has life in it

The IBM Power high-availability server was upgraded with parts from the previous production server that could still be put to good use. Renewing doesn't always mean replacing everything.

4. Add high availability where there is none

A database on a single server is a single point of failure, however well it is run. With a dedicated node under VMware Enterprise Plus and monitoring through vRealize Operations, the database gained the software-based high-availability strategy the cooperative needed.

5. Plan backup by platform

IBM i and VMware have different backup tools, and it pays to use the right one for each. BRMS is IBM's native solution for managing backup, recovery and media on IBM i; Veeam is one of the most widely used data protection platforms for virtualized environments. Both write to IBM TS4300 tape libraries, which makes it possible to keep offline copies as part of the protection strategy.

The result: more capacity, less complexity

The cooperative gained unified storage for both environments managed from a single console, high availability for its database node, IBM Power9 capacity for its core migration and a renewed backup scheme on both platforms. All with current-generation equipment, such as FlashSystem, that meets or exceeds its requirements, and without disrupting critical operations.

Where to start

If your institution has platforms near their capacity limits, databases without high availability or a core migration underway, it's worth looking at the full picture before buying piece by piece: capacity, storage, backup and licensing. At Redsis we combine more than 25 years of infrastructure experience in Latin America's financial sector with a portfolio that brings together IBM, Lenovo, VMware and Veeam.