In this section, we'll cover key goals to always keep in mind when designing any system:
When you prepare food, no matter the ingredients or recipe, you want it to be tasty, healthy, and affordable.
Just like these qualities make a great meal, these goals make a great system.
Imagine a local bakery that initially handles its customers with just one cashier.
Now the bakery becomes more popular.
Because of that, the line of customers grows longer, and waiting times increase.
To serve more customers, the bakery opens additional checkout counters, helping them handle the growing crowd efficiently.
Basically, our bakery has scaled up to meet increased demand. Just like the bakery, our technical systems also need to scale up as more users join.
A well-designed system can scale up more easily.
Scalability is the system's ability to handle more work smoothly as demand grows.
Availability means how much time a system is up or operational.
For example, an online banking website that is available 24/7 ensures that users can access their accounts and perform transactions at any time.
A system that is available 99.999% of the time, also known as "five nines," means it is only allowed 5 minutes (0.001%) of downtime per year.
Consistency means consistent/same data visible to everyone.
For example, if you update your profile picture, every user sees the updated picture, and nobody sees the old one.
Ensuring consistency is important for applications where we need up-to-date information. One good example is financial transactions.
When you withdraw money from the bank, it’s essential that the updated balance is immediately reflected so the same money isn't withdrawn multiple times.
Strong Consistency means consistent/same data visible to everyone instantly, without any delay.
For example, when withdrawing money from the bank, it’s essential that the updated balance is immediately reflected so the same money isn't withdrawn multiple times.
Eventual Consistency means data will become consistent over time, but not necessarily instantly.
For example, if you update your profile picture, some users might still see the old picture for a short while, but eventually, everyone will see the updated one.
Eventual consistency can be used on social media or in applications where there is no strong need for instant updates.
Let's assume a very simple system where a user is trying to open sweetcodey.com.
There is a server that handles the client’s request and a data store that keeps the site data.
We can clearly see that if this server goes down, the website will be inaccessible. Here, this server is a Single Point of Failure (SPOF).
SPOF (Single Point of Failure) is a component in a system that, if it fails, will stop the entire system from working.
We can avoid SPOF in our system by adding more servers (aka adding redundancy).
This would make our system more tolerant to failures, as it will continue to operate even if some parts or components fail.
Technically, we say that we have made our system Fault Tolerant.
- If one server in the data center fails, another server takes over.
- If one data center fails, another data center takes over.