Less than 1000 requests per second in peak without TLS

or

Less than 700 requests per second in peak with TLS enabled

Run LR on the core node (assuming that other services are not overloading the core node).

More than 1000 requests per second in peak without TLS

or

More than 700 requests per second in peak with TLS enabled

Install a separate node for LR.

LR node requires one core 2.2GHz per 1000 requests without TLS (or equivalent) or 700 requests with TLS enabled.

Set the logreceiver.verticles_count property to the number of cores of the machine. You may double this number if your cores support hyperthreading. For information, see Setting verticle scaling of log receiver.

Recommended LR memory is 4GB for up to 16 cores, 6GB for 32 cores, and 8GB for 64 cores.

If vertical scalability cannot handle the LR load

Set up multiple LR nodes with the load balancer. See Configuring load balancing for Log Receiver (LR).