How to Optimize InfluxDB Performance for Large Time Series Data Sets?

1

Hello guys :smiling_face_with_three_hearts:

I am working on a project that involves storing and querying large time series datasets in InfluxDB, and I have need some advice on how to optimize performance.

Specifically, I have a issues these area: :pensive:

Schema Design: What are the best practices for schema design to ensure efficient querying and storage?

Retention Policies: How should I structure my retention policies to balance between data availability and storage efficiency?

Hardware Recommendations: What are the recommended hardware specs (CPU, RAM, Disk I/O) for handling large datasets in InfluxDB?

Query Optimization: What techniques or strategies can I use to optimize query performance?

Maintenance Tips: Any advice on regular maintenance tasks that can help keep the database running smoothly?

I also check this article :point_right: optimizing InfluxDB performance for high velocity dataqlik but I have not found any solution. I’m using InfluxDB version 2.x and any version-specific tips would be greatly appreciated.

Thanks :innocent:

Respected community member
Deniz Can :smiling_face_with_three_hearts:

1 Like
2

Just created an account to join in on this question.
As I need to setup a time series database / cluster to handle /store measurements.

At the minute when I test it in vanilla out of the box influx 2.7 it gets unresponsive after about storing 50gb of data for about 40k measurements writing a record every 30 seconds. It also won’t startup again at this point.

If anyone has pointers where to look for tips on setting up influxDB 2.7 for larger amounts of measurements that would be lovely.