3 ms·
This reads like it was written by someone who has never worked with a properly managed distributed system. What I can believe is that it's much easier to misma
by joatmon-snoo 7y ago
This reads like it was written by someone who has never worked with a properly managed distributed system.
What I can believe is that it's much easier to mismanage a distributed fleet of consumer PCs, as a result of mainframes mostly only having been used in situations with extensive change management and change audit processes.
- XorNot 7y agoI'm curious how, with presumably exactly 1 large capex asset, changes are actually tested or even developed.
- RantyDave 7y agoEverybody has a test environment. Some people are lucky and have an entirely separate production environment, too.
- Spooky23 7y agoIt’s all little VMs, which are exponentially more expensive and significantly less capable than a normal computer.
- rbanffy 7y ago> significantly less capable than a normal computer less capable than the underlying mainframe, you mean. Unless you have tons of separate VMs running on your mainframe, you can have tens of terabytes of RAM available to your job and a fridge-sized cabinet of PCIe channels to do its IO (which is done by specialized CPUs).
- rbanffy 7y agoMainframe environments are partitioned. Usually, when you want to validate a change, you deploy it to a separate partition of the same machine that is running your current workload. This is a huge simplification.