2 ms·
As a data point, at my current place (a large, university-affiliated R&D lab), the upgrade cycles are different for "servers" that do some externally visible /
by ptero 10y ago
As a data point, at my current place (a large, university-affiliated R&D lab), the upgrade cycles are different for "servers" that do some externally visible / useful stuff and workstations.
Servers generally get upgraded when the next functionality upgrade topples the current hardware or when the hardware physically dies (often WELL over 5 years).
Workstations / laptops are upgraded whenever a user wants a new one. Many folks (myself included) prefer to keep our current systems for a long time. Components and peripherals get upgraded much more frequently, as people keep buying hard drives, GPUs and other new toys.
- tyingq 10y agoYes...a fixed refresh cycle is more common in corporate environments. And it's more of a budgetary thing. IT has some flexibility to keep old servers, but the budget they have to refresh is calculated on some X years cycle. So they are limited in the number they can purchase each year to refresh existing servers. Net new functionality or applications typically fund the first cycle for servers that have a net new use.