[System Design] Scale From Zero To Millions Of Users

[reboottime]

1. How Web browser, DNS server, and web server work together

2. RDMS VS NoSQL

• Types: key-value stores, graph stores, column stores, and document stores
• when you may consider using NoSQL
  • Your application requires super-low latency.
  • Your data are unstructured, or you do not have any relational data.
  • You only need to serialize and deserialize data (JSON, XML, YAML, etc.).
  • You need to store a massive amount of data.

Vertical scaling vs horizontal scaling

• Vertical scaling: use better resources
• horizontal scaling: user more resources

Horizontal scaling is more desirable for large-scale applications due to the limitations of vertical scaling.

• single point failure
• limitation on resource limitation

[reboottime]

Load balancer, a horizontal scaling solution at web tier

[reboottime]

Database replicas, a database tier scaling solution

High traffic read or write or both read and write could happen on database component in the whole web application system

The solution to mitgate the database bottom neck is using database replicas.

• master database: A master database generally only supports write operations
• slave database: A slave database gets copies of the data from the master database and only supports read operations.

Two complicated questions

• what if the master database is down?
• what if all the slave databases are down?

[reboottime]

Cache

A cache is a temporary storage area that stores the result of expensive responses or frequently accessed data in memory so that subsequent requests are served more quickly.

• When to use
• what are the challenges
  • data inconsistency

[reboottime]

CDN

[reboottime]

So far, the whole application's high level view looks in this way

[reboottime]

Further improve the application scaling

Move stateful data outside of the web tier.

[reboottime]

Further scaling up

1. Data Center:
   • geoDNS is a DNS service that allows domain names to be resolved to IP addresses based on the location of a user.
2. Message Queue: A message queue is a durable component, stored in memory, that supports asynchronous communication.
3. database sharding
   • sharding key

[reboottime]

References

All the resources are from https://bytebytego.com/courses/system-design-interview/scale-from-zero-to-millions-of-users