Hi
This is a more theoretical question, let's say we have customer with offices in four European countries. Total number of concurrent connected devices are 15000. Latency and communication speed well within limits.
One country as the main country there two N3000 servers are placed with the majority of the users. Two countries have about 2000 concurrent connections and one N3000 server and the first country as secondary Radius server.
The last country have only 500 concurrent connection. Based on the limits of the concurrent sessions for N1000 this hardware would be enough for this country. But in the Scaling and Ordering Guide it's mentioned a maximum of 4000 Endpoints in the database.
![](https://higherlogicdownload.s3.amazonaws.com/HPE/MessageImages/e997033eaefd4f8eb2388ebd25e89194.png)
If this server is a Subscriber in the same cluster it will get all the endpoints from the other countries, would this be a problem?
I haven't noticed the same figures before in older versions of the document. But maybe just missed them.
------------------------------
Best Regards
Jonas Hammarbäck
MVP Guru 2024, ACEX, ACDX #1600, ACCX #1335, ACX-Network Security, Aruba SME, ACMP, ACSA
Aranya AB
If you find my answer useful, consider giving kudos and/or mark as solution
------------------------------