When I’m talking to buyers, customers, partners, and analysts, I typically summarize all of the data center trends using the chart above. It’s from IDC’s Digital Universe Study – and shows over a ten year period a 50% growth in IT professionals compared to a 10X growth in virtual and physical servers and a 50X growth in data with a 75X growth in objects.
One interpretation of this chart is that if you have an IT group of two (including yourself) today with 20 virtual and physical servers and 10TB of data, that in ten years (with flat revenue) you can project that you’ll have a staff of three who will be managing 200 virtual and physical servers and one petabyte of data.
This is what leads to the title of the slide: “adapt or be crushed” (I like “adapt or die” better, but have been told it’s a bit too hard core.) This adaptation is what enables order of magnitude changes in how many servers, how much data, and the like that IT staff can handle. In short, this adaptation is what the larger data centers are doing today – with their ratios of 20,000 physical servers to one administrator/operations person. In essence, it comes down to building agile, automated, adaptive data centers.
We’ll discuss the characteristics of data center agility, automation, and adaptation in the next post in this series.
This is part 6 of an on-going series. Part 1: virtualization isn’t the next big thing (NBT) because it was the last big thing (LBT); part 2: data center IP traffic growth; part 3: data center IP traffic sources; part 4: cloud workloads; part 5: large data centers and administrator to server ratios.