180 points by dataengineer 6 months ago flag hide 10 comments
db_enthusiast 6 months ago next
Great work! I've been looking for a time-series database with built-in anomaly detection and forecasting capabilities. I'll definitely give this a try in our AI-driven analysis tool.
op 6 months ago prev next
Thank you! We're looking forward to feedback from the community on the project and will be glad to integrate enhancements based on user needs.
ml_obsessed 6 months ago prev next
How does this stand out from other open-source time-series databases such as InfluxDB and Prometheus?
op 6 months ago next
Both InfluxDB and Prometheus are exceptional projects. However, where we think our solution stands out is by providing an integrated experience for forecasting and anomaly detection, which usually require separate tools and integrations. In addition, we provide a flexible and easy-to-use query language that simplifies data analysis.
scale_fanatic 6 months ago prev next
I'm concerned about the level of scalability offered by your solution. How does it handle handling large time-series datasets?
op 6 months ago next
Our solution has been designed to handle large-scale datasets by leveraging column-store and horizontally partitioned data structures. We'd be more than happy to discuss our design decisions and how we scale in more detail if you're interested!
fp_engineer 6 months ago prev next
I immediately checked out the project after reading your post. GitHub tells me your repository has more than 8k stars! Congratulations and thank you for creating this resource for the community.
op 6 months ago prev next
Thank you for your support, fp_engineer! We're excited to receive continuous contributions from the community to make our system even more efficient and feature-packed.
crypto_analyst 6 months ago prev next
This is a fantastic resource for people into DeFi and blockchain. I'm curious as to what factors you considered before you went ahead and used Rust as the pile of your choice?
op 6 months ago next
Rust was the optimal fit for us given its strong memory safety guarantees and performance characteristics, which are crucial for time-series databases. We're always monitoring the developments in the open-source ecosystem and our choice of tools might evolve as new technologies emerge.