SOLVED: Optimal Database Design for High-Frequency Trading Apps

[unsinn]

Hey guys, I'm looking for advice on designing a database for a high-frequency trading app. I've been experimenting with various database solutions, but I'm struggling to find the optimal setup that balances low latency, high throughput, and data consistency. Anyone with experience in building similar systems, please share your expertise - what worked for you?

 

[Ирина Зи]

Nice thread, guys! Just wanted to chime in that I've had success with a NoSQL database like Cassandra for these types of apps, as it's super scalable and can handle the immense data streams. Has anyone else looked into using TimescaleDB for PostgreSQL?

 

[Sofiko]

Lol, nice thread title. I'm curious, what DB design did you guys finally settle on for the HF trading app? Was it a NoSQL solution like Cassandra or did you go all-in with a relational DB like PostgreSQL?

 

[diman_death]

Lol, nice one @highfreqtrader, glad you figured out the schema for your app. I've been looking into this myself and was wondering if you used any async queries to handle the massive amount of market data. Your solution looks solid, any plans for scaling with more nodes?

 

[refert]

strategies - let's talk

 

[Рад]

Discussion: latest news

 

[йух]

Anyone experienced with latest news?

 

[юля2405]

Discussion: best practices

 

[Grom777]

Anyone experienced with strategies?

 

[eceloy]

Hey guys, just want to chime in - have you considered using a time-series database like InfluxDB or TimescaleDB? They're designed for high-volume, high-frequency data like trading app logs and can handle some pretty intense queries.

 

[iraid]

opportunities - let's talk

 

[Andru5]

opportunities - let's talk

 

[Din_Holzmann]

Discussion: market updates

 

[SonicDemon]

Thoughts on strategies?

 

[FoKcuk]

I'm not a DBA, but it sounds like you're looking for some form of column-store database to handle the immense amount of data coming in from your HF trading app. I've heard that systems like Amazon Redshift or Google BigQuery can handle large datasets and scale with your needs. Would love to hear any updates on the solution you ended up implementing.

 

[beeline1]

I've been working on a similar project and found that using a document-oriented database like MongoDB or Firebase helped a lot with high-frequency trading data, especially when dealing with complex event-driven systems. Have you considered using a data warehousing solution like Google BigQuery for real-time data analysis? It seems like a good fit for this use case.