So I said we'd come back to that SE thing with the keyboard, screen, and the servers, the computer within a computer. What's that there fore=? That's the support element or the SE. In order to get the mainframe all configured and ready to boot up the way we want it to and connected everything we needed to talk to, that takes a little bit of a setup, and you do that setup through the support element. There's two SE's in there, and that's just in case one of them fails. No one wants to be locked out of their system. Now that's all cool, but imagine if every time we needed to do something, you had to walk over to the data center, badge in, hang out on the raised floor where it's all loud, and cold, and dark, and stuff, open the door, slide out the SE, and all that, just to type out a single command. Luckily, in addition to the SE's, mainframes have this concept of the HMC or the hardware management console, and these do four things. One, they let you access the SE's Functions remotely, so you don't have to walk out there anymore. Two, you can add multiple mainframes to a single HMC, a 100 of them in fact. So you can manage a whole bunch of mainframes from one HMC and that is pretty handy. You can also add a single mainframe to multiple HMC's, 32 of them in fact. So if you have one system that's shared by multiple departments, they can all get HMC access. Four, you can connect to the HMC remotely. So while the SE is attached to the system and the HMC is running somewhere else, you can also connect remotely as long as you have the correct network access. Now for the remainder of this session, I'm just going to be saying HMC because that's typically how someone accesses those functions. But know that pretty much anything I can do from the HMC, I can do from the SE as well. It's just a whole lot easier just to say HMC. So why would we need that kind of access? Well, the most common thing people use an HMC for is to stop, start, create, and modify our pars. Need to add another couple processors to an L part, you do that from the HMC. When it changed the way a network card is configured, you do that for the HMC. Let's say I'm having problems with the system and I want to get some data that might help me with troubleshooting, guess what? HMC. What do I see when I login to an HMC? Well, I can take a look at a particular physical machine that's this view right here. See you got the status is operating, It's going to machine type, model number, It's also got a serial number, It's also got this name over here, so I can make sure it's the system that I want to be talking to. If I go deeper, I can see all the L parts, those are the partitions right there for that system. Looks like some of them are up, some of them are down. That's all right. In this last tab over here that's the topology, that shows me how everything is connected. You can see there's lots of stuff going on over here. Another good use of the HMC is just telling at a glance how my systems are doing. So I can see right here there's a couple of exceptions. An exception is an unacceptable status and it could be a problem with lpar or any other objects like a network or a crypto card. These need to be checked out right away. There's also a number of hardware messages. Now, a hardware message is just when some components saw something and at once you don't know about it, but it's not currently stopping anything from working, it's just letting you now. For example, if a fiber cable got disconnected and then reconnected, that would result in a hardware message. Same thing if an lpar didn't shut down correctly or HMC somewhere lost connectivity, that would result in a hardware message. So what gets stored on the SC? Well, it's got code that's going to get loaded onto the mainframe, that's called Licensed Internal Code or LIC. Yeah. Really. We've got logging and problem determination Functions which help in debugging issues, hardware systems definitions which basically tell the mainframe how to get started up when it first fluids up. Then there's the IOCDS which will definitely be talking about later on. That tells the mainframe how to configure all of its IO and it's L pars. There's also a battery operated clock that it uses a synchronized time across all of the components and Ethernet adapters because of course everything in the mainframe environment generally is connected to other components, so we need that. Lastly, you might be thinking, so if anyone could log into the HMC, they could probably mess up a whole bunch of stuff accidentally. Well, which is why there are several accounts set up by default on the typical HMC and most people set up even more to partition access off to those who really need it. Just because you have access to some features, doesn't mean you have access to everything. There's specific roles on the HMC, roles like operator, and Advanced operator, Systems programmer, access Administrator, and service representative. Each one of these roles has a specific reason for needing specific access to the system, so it's not an all or nothing kind of thing. That's the SE and the HMC, the computers within your computer, and within those computers, there's actually no there's not. That's it. It's just the SE and the HMC.
