We are currently working on moving from on premise to hosted data centers using modern infrastructure. Internally, there is a lot of discussion around how often our technologists will need to visit the data centers to perform work once the data centers are in service. Does your organization keep metrics of how often your techs need to visit your data centers and are willing to share?

3.9k viewscircle icon5 Comments
Sort by:
Manager, Network Engineering3 months ago

When we were utilizing hosted data centers we visited approximately 2 days per month.   

Director of IT3 months ago

Happy to share some insights from our experience on this topic.

We transitioned from on-premise to hosted data centers over the past 18 months in my previous job. As part of that shift, we significantly reduced the need for on-site tech interventions thanks to improved remote management capabilities.

🔹 Average visits: After stabilization, we’ve averaged 1 physical visit every 2-3 months, typically for hardware audits, vendor coordination, or network upgrades.

🔹 Remote-first strategy: 95% of operational tasks—monitoring, patching, provisioning—are handled remotely via secure management consoles.

🔹 Key enabler: Implementing a robust IAC (Infrastructure as Code) and remote access control strategy was essential to minimize visits.

Hope this helps the community benchmark their own approach.

Lightbulb on1 circle icon1 Reply
no title3 months ago

Thank you very much for the response. Very helpful.

General Manager - Technology3 months ago

I have hundreds of servers in multiple Datacenters . We make sure we have an agreement in place with the respective Datacenters , regarding routine visual inspections, Documentation, security protocols, and a hand and feet service. Effectively we never need to visit a Datacenter, even when removing an old server or installing a new host. 

Director of IT in Healthcare and Biotech3 months ago

We use remote access tools to access and maintain our systems in our cloud hosted environments.  You should shop for a hosting provider that provided this capability.

Content you might like

Read More Comments

Yes, we do today.10%

No, but we plan to in the next 6 months.34%

No, but we plan to further in the future.10%

No, and we have no plans to.44%

View Results

Organisation adopts the CSP-provided OS images for services like AWS - ECS, EKS, GPU, EMR, GCP - Dataproc ( due to agents required to work with CSP control plane )38%

Organisation hardens CSP-provided OS images according to own standards58%

Organisation does not harden CSP-provided OS images and relies on hardening done by CSP35%

Organisation does not adopt CSP-provided OS. ( Not sure how organisation would use more complex services such as EKS, EMR, ECS, without challenge of baking in agents )23%

For simple compute scenarios the organisation uses OS image consistent with private cloud and either brings own image or bases off CSP marketplace 13%

Organisation does not look to maintain an OS image consistent with private cloud8%

View Results