I want to know Best Practices for Designing an Efficient Data Model in ScyllaDB?

ScyllaDB
, ,
1

Hey everyone,

I am working on designing a data model for a new project using ScyllaDB & want to get some insights from the community. I know that ScyllaDB is built for high performance & scalability but I want to make sure I am structuring my tables and queries in the best possible way.

From what I understand, avoiding too many partitions and making good use of clustering keys is key But I am still a bit unsure about things like

  • How do you decide on the right partition key strategy for high write workloads?
  • Any tips on avoiding tombstones and keeping read performance fast?
  • How do secondary indexes and materialized views impact performance in real-world scenarios?
  • Are there any anti-patterns you wish you had avoided when designing your data model?

Would love to hear from folks who have worked with ScyllaDB in production.

if anyone have any resources, tutorials or personal experiences please share with me.

Thank you……. :slight_smile:

2

There are many tutorials and lessons, covering the topics you mentioned in ScyllaDB University.
The How to Write Better Apps lesson covers best practices and tips, including common pitfalls, performance, and tombstones.
Materialized Views Indexing and Filtering, including performance impact, is covered in [this lesson](https://Materialized Views, Secondary Indexes, and Filtering).
If you have specific questions, you can ask them here. Provide as many details as possible about your use case and the data model.