Grokking System Design Fundamentals
Ask Author
Back to course home

0% completed

Vote For New Content
The lone Ranger
Confused??

The lone Ranger

Mar 16, 2024

2. Scalability

Evaluating your application’s scalability needs, both in terms of data volume and read/write load, is crucial in choosing the right database.

SQL databases: SQL databases are known for their vertical scalability, which means adding more resources (CPU, RAM, storage) to a single server to handle increased workload. This can be suitable for applications with moderate scaling requirements, such as small to medium-sized web applications or internal company tools. However, this approach can be expensive and has limitations as the server's capacity is finite.

NoSQL databases: NoSQL databases provide horizontal scalability, allowing you to distribute data across multiple servers, making it easier to handle large volumes of data or high traffic loads. This is advantageous for large-scale applications with high throughput and data volume requirements, such as big data analytics, real-time data processing, or Internet of Things (IoT) applications. If your application requires easy scaling to accommodate growing data or user bases, NoSQL databases are a better choice.

In above paragraph why its written that SQL db is only vertically scalable? Can't we add more servers and scale horizontally? And there is a db called amazon aurora rds which scales automatically. So I think this paragraph needs improvement sir. Please check and resolve.

0

0

Comments
Comments
Design Gurus
Design Gurus2 years ago

SQL databases are 'known for their vertical scalability'. They can still be scaled horizontally, but it is a difficult process and there are many challenges compared to NoSQL databases.

See this blog for the challenges of scaling SQL databases: [https://medium.com/gi...

Chris Fitzpatrick
Chris Fitzpatricka month ago

The talking points on NoSQL (particularly around scaling) read like they were written in 2009, not 2025. The disparity in horizontal scaling ability between relational and NoSQL DBs is nowhere near the extent the author makes it out to be, which is really disappointing ...

On this page