Reliability in computer systems
Introduction
Computers need to be reliable for people to use them and have confidence in them. If they are not reliable, then they will quickly be abandoned.
Reliable
Imagine when you tried to use your phone that it never worked in the same way each time you used it, or that it kept breaking down. Imagine that text messages got through to you only every now and again. How would that make you feel? What would you do? Imagine the remote control for your TV worked only occasionally or your computer kept breaking down and you had to keep sending it off to be repaired. Imagine the computer controlled microwave never cooked your food properly! It would be a very frustrating experience.
-
- You might stop using the equipment for good.
- You might get very stressed out and frustrated over time.
- You might decide never to buy a product from that manufacturer again.
- Manufacturers might go out of business if enough people felt the same way.
- You might change your habits.
Computer systems need to be reliable if they are to be confidently used. If people cannot rely on them, then they will quickly be abandoned.
Failover
If an organisation's network or key pieces of hardware fail for whatever reason, it might have serious implications for the organisation. They might even go out of business. If some of the network hardware failed on your school or college network, you might lose all of your controlled assessments or other project work. The school might lose all of its information it needs to run the school and information about you, for example, contact information in case you have an accident.
All equipment will fail eventually. Moving parts wear out. Dust gets into equipment and stops them working properly. There are other reasons, too. So that a business can continue as normal in the event of a serious hardware failure, organisations often use 'failover'. Failover is the term used when you have a second, identical piece of equipment that can start working automatically if the first main piece of equipment fails.
For example, one key piece of equipment on a client-server network is the server. This manages the entire network of personal computers, printers and other hardware on the network. If the server fails, the whole network will not be available for anyone to use.
Network designers often include a second, back-up server on a network. This second server has a mirror image of the software and files on the main server, and is constantly updated. If the main server fails, then this second server starts up automatically so that users on the network can carry on as normal. They won’t even know that there has been a major hardware problem. The network manager can then arrange to get the main server fixed.
You could ask your Network Manager if they have a back-up server that kicks in automatically in case the first one fails. You could also ask about the back-up system in place. Are magnetic tapes used? How is the back-up system organised? How often are back-ups taken? Have they considered using cloud storage for automatic back-ups if they don't? Why do they / don't they use cloud storage?