Edge BYOC WebSocket drops at 1500 concurrent users on US1

Ran into a weird issue today with the Edge BYOC WebSocket configuration during our latest capacity planning tests. We are trying to validate the maximum concurrent session limit for our inbound voice traffic on the US1 environment. The goal is to push the system to its documented limits without triggering premature connection drops.

We are using JMeter 5.6.2 for the load generation. The script simulates a steady ramp-up of WebSocket connections to the Edge BYOC endpoint. Everything works fine up to about 1200 concurrent users. The handshake completes, the SIP INVITE is sent, and the media stream establishes. However, once we cross the 1500 user threshold, we start seeing a high volume of 1006 (Abnormal Closure) errors. The connections are not being rejected by the API (no 429s here), but they are failing silently after the initial handshake.

Here is the relevant JMeter thread group configuration:

thread_group:
 name: BYOC_Load_Test
 number_of_threads: 2000
 ramp_up: 600
 loop_count: 1
websocket:
 endpoint: wss://us1-platform.genesyscloud.com/v2/interactions/edge/byoc
 headers:
 Authorization: Bearer ${access_token}
 X-Genesys-Application-Id: ${app_id}
 X-Genesys-Application-Version: 1.0.0
 keep_alive: true
 timeout: 30000

The error logs show that the WebSocket frame is sent, but the server closes the connection immediately after the SIP 200 OK is received from the downstream endpoint. This happens consistently at the 1500 mark. We have verified that the API token is valid and has the correct scopes (interaction:initiate:agentless).

Is there a specific limit on the Edge BYOC WebSocket connections per application ID or per tenant that we are missing? The documentation mentions general capacity limits, but it is not clear if there is a hard cap on concurrent WebSocket connections for BYOC. We need to understand if this is a configuration issue or a platform limitation. Any insights into the expected behavior at this scale would be helpful.