Hello ,
I'm currently evaluating Traccar as a solution for my GPS tracking needs. I have a specific use case where I expect to have a high volume of GPS devices connected to the Traccar server—approximately 3 million devices.
What is the maximum number of devices that Traccar can efficiently handle?
Are there any built-in limitations or performance bottlenecks that I should be aware of?
Has anyone successfully scaled Traccar to support around 3 million devices? If so, could you share some insights or tips on how you achieved this?
Are there any additional infrastructure considerations for scaling to this level, such as load balancing or database partitioning?
I appreciate any insights you can provide on this matter. Thank you!
To answer your question about data frequency, each device will be sending data at 15-minute intervals.
I have further questions about Traccar's horizontal scaling capabilities:
For instance, if I want to scale out to 3 Traccar instances:
- Data Synchronization: How should these instances be configured to effectively synchronize data?
- I've read through the documentation and noticed that multicast messages might be the way to go. Should I configure Instance 1 to synchronize with Instance 2, and then configure Instance 2 to synchronize with Instance 3?
- Alternatively, should Instance 1 be configured to synchronize directly with both Instance 2 and Instance 3 by reusing the 'broadcast.address' configuration multiple times?
- Device Grouping: Would it be advisable to designate specific server instances to handle data from specific groups of devices? In other words, should I configure each group of devices to report their data to a specific instance's IP address?
You should probably read the documentation first.
I'm bad at coding but at infrastructure I defend myself and if you don't have a battery of servers, with so many internet and "SIM" GPRS connections in reception, your project is doomed to failure and I'm talking about 10 million in investment, everything else will be DIY, because yes traccar is solid, but the flows are enormous
Why I'm not seeing an issue?
You could probably DNS round-robin multiple instances and cluster the database - depends on current load - Sharding is not needed.
You might just need to handle/rotate the logs periodically.
For cost estimation I belive you could push fake data to a test cloud setup then predict the amount.
Hello ,
I'm currently evaluating Traccar as a solution for my GPS tracking needs. I have a specific use case where I expect to have a high volume of GPS devices connected to the Traccar server—approximately 3 million devices.
What is the maximum number of devices that Traccar can efficiently handle?
Are there any built-in limitations or performance bottlenecks that I should be aware of?
Has anyone successfully scaled Traccar to support around 3 million devices? If so, could you share some insights or tips on how you achieved this?
Are there any additional infrastructure considerations for scaling to this level, such as load balancing or database partitioning?
I appreciate any insights you can provide on this matter. Thank you!