Why Do Mainframes Still Exist? What's Inside One? 40TB, 200+ Cores, AI, and more!

"Why Do Mainframes Still Exist? What's Inside One? 40TB, 200+ Cores, AI, and more! - Dave explores the IBM z16 mainframe from design to assembly and testing. What's inside a modern IBM z16 mainframe that makes it relevant today?" - by Dave Plummer. This is an amazing 23-minute video by the Microsoft programmer who developed the Windows NT Task Manager, among other things. He visits IBM and talks to engineers about the Telum chip architecture (Hot Chips 2023), used in the z16 mainframe. Special attention is paid to the cache. ▶ this video

Check out this youtube video: ▶ this video It does a reasonably good job at comparing mainframes to a regular server setup. It's not about SPEC or GFLOPs etc.

Dave Plummer recently did a tour: ▶ this video It's definitely intriguing if it were.. I don't know, available to tinker with?

Great comment. IBM is as IBM wants to be. I don't think there's a mystery to it and they've ended up on an evolutionary path that can't be reversed. To wit, I just watched a long form video [1] on the passion that goes into IBM mainframes including some interesting clips of various physical robustness tests of the assembled systems. It reminded me of watching clips from various space programs where a largely full-up untested rocket had to put humans into orbit without blowing up. Another revolution that's occurred, but IBM chose not to really follow (asterisk) is the idea that you assume some percentage of your hardware will fail (instead of trying to make it never die) -- then you engineer around that. I'm not sure what will win in the end, but right now the entire industry is telling us to have failover into different availability zones in Amazon, or even across clouds -- one of which IBM offers. 1 - ▶ this video (asterisk) - IBM does sort of have an understanding of this of course, they have a cloud after all.