In the before times, if a company had a website and internal services like a sales system or a production management system they would have to have a physical server that has enough performance to run these services simultaneously at peak times, and enough storage to hold all the required data.
These servers weren’t cheap. They had expensive hardware, required special backup power systems, you had to arrange for redundant systems and offsite data backup, along with the power and cooling that big ol’ computers require
“The Cloud” is just another word for “Someone else’s computer”. Amazon, Microsoft, and Google all maintain some big data centers scattered around both the world and the country. This scattering of datacenters provides great redundancy, you want backups in case there’s data corruption, but scattering datacenters around means a big storm in Tennesee shouldn’t take out your system because Michigan has copies of the same data. The cloud providers also aren’t buying a few dozen hard drives from CDW for their servers, they’re buying millions of hard drives direct from the manufactures which offers them good savings on each drive. Then they’re building a big empty building with a single large backup power system rather than each server needing its own little one. They’re also grouping together tons of customers onto the same machines so each server is always staying busy, and if the main customers aren’t using all the CPU time then they can just sell off the CPU time to others who need a lot of hours, but not high priority so their tasks can just fill in any gaps that show up.
The end result is that its cheaper per TB of storage and per hour of computer time to run a huuuuugeeee number of servers than for each company to run their own so all the companies took their former server budget and shifted a portion to getting cloud services instead
Latest Answers