Home›Insights›Articles›Delegated IT infrastructure management: running 24/7 with no engineers on site
Blog · Managed services · Infrastructure
Delegated IT infrastructure management: running 24/7 with no engineers on site
Hundreds of virtual machines, more than 300 databases and two data centers, all run 100% remotely. What we learned managing the infrastructure of a Big Four professional services firm in Latin America.
The size of the job
The service scope gives a sense of what it takes to run a modern corporate infrastructure:
- 484 VMware virtual machines , 369 of them in production.
- 9 physical x86 servers split between the primary site and the disaster-recovery (DR) site.
- 5 IBM FlashSystem storage arrays .
- 300+ SQL Server and MongoDB databases under management.
Add a multi-platform environment — VMware, Windows, Linux, IBM Power and IBM FlashSystem — that calls for different specialists at every layer. Covering all of that with in-house staff, in 24-hour shifts, 365 days a year, takes a sizable team that's hard to retain.
What "delegated management" really means
Delegating infrastructure isn't the same as outsourcing tech support. In a delegated-management model, the provider owns day-to-day operations: it monitors, detects, responds, escalates, documents and reports. The client keeps strategic control and visibility but no longer carries the operational load.
To work, the model needs four pieces:
- A permanent operations center watching the infrastructure 24 hours a day.
- Standardized processes , ideally built on recognized frameworks such as ITIL.
- Criticality-based SLAs , so each incident gets the attention its impact demands.
- Traceability : the client must be able to see what happened, when, and what was done.
Delegated management works when the client knows exactly what's happening in its infrastructure — without having to run it.
How we run it: the model behind the service
For this client, Redsis delivers the service through its Customer Service Center (CAC), which follows ITIL practices with 24/7/365 monitoring. Incidents are classified by criticality, and Level 1 incidents carry a 15-minute response time.
Criticality-based classification is a core ITIL practice: an incident's priority comes from combining its impact — how many users or services it affects — with its urgency — how fast the damage grows. A downed test server doesn't weigh the same as an unreachable production database, and the SLA should reflect that.
The entire service is 100% remote, with no engineers on site. The client has access to a self-service portal and receives monthly management reports, giving it full traceability over operations.
Five lessons for delegating infrastructure without losing control
1. Take inventory before you sign
A good managed service starts with an accurate inventory: how many virtual machines, how many in production, which servers, what storage, how many databases and on which engines. Without that baseline, SLAs get negotiated in the dark.
2. Set criticality with the business
IT shouldn't define the priority matrix alone. Business units need to agree on which services are critical and what each one's impact is. That agreement is what makes it possible to commit to, say, a 15-minute response for Level 1 incidents.
3. Treat DR as part of operations
A disaster-recovery site isn't an insurance policy you check once a year. For this client, coverage includes both the primary and DR sites, so business continuity is part of the daily service — not a plan sitting in a drawer.
4. Put platform capabilities to work
Modern storage does part of the job. IBM FlashSystem, for example, offers AI-based ransomware detection, immutable copies (Safeguarded Copy), data replication and centralized management across systems through IBM Storage Virtualize. An operator that knows those capabilities inside out can use them to strengthen protection and continuity.
5. Measure — and show it
An SLA that isn't reported doesn't exist. The self-service portal and monthly management reports make operations auditable: the client can verify compliance, spot trends and make informed decisions about its infrastructure.
The outcome: mission-critical infrastructure at a lower operating cost
Today the firm has 24/7/365 infrastructure operations under ITIL, criticality-based SLAs with a 15-minute response for Level 1 incidents, coverage of its primary and DR sites and an end-to-end managed multi-platform environment. All of it with no engineers on site, at a lower operating cost than a dedicated in-house team, and with full traceability through the self-service portal and monthly reports.
For Redsis, this engagement proves out something important: a delegated-management model that is scalable and repeatable for large enterprise accounts, able to run mission-critical infrastructure under auditable SLAs.
Where to start
If your IT team spends most of its time keeping infrastructure running, if covering nights and weekends is a recurring headache, or if you lack a clear view of whether you're meeting your own service levels, it's worth evaluating a delegated-management model. At Redsis, we bring more than 25 years of infrastructure experience in the region, IBM Platinum and Cisco Premier partnerships, and a Customer Service Center that runs under ITIL around the clock.
Read the full story
See how Redsis runs the IT infrastructure of a Big Four professional services firm.