Rate limiting mechanism based on device load/capacity or traffic content

A network device having a symmetric multiprocessing architecture, the network device comprising:

• a plurality of CPU cores;
• a network interface to receive one or more data packets;
• and
• a memory whose access is shared by the dedicated CPU core and the plurality of CPU cores;
• wherein the plurality of CPU cores are to:
• receive a request from a network node;
• determine whether a ratio between a current load of the network device and a capacity of the network device exceeds a threshold;
• and
• in response to the ratio between the current load and the capacity exceeding the threshold,
• determine a wait time period based on the ratio between the current load and the capacity;
• and
• respond to the network node with a message including the wait time period.

A non-transitory computer-readable storage medium storing embedded instructions for a plurality of operations that are executed by one or more mechanisms implemented within a network device having a symmetric multiprocessing architecture, the plurality of operations comprising:

• receiving a request from a network node;
• inspecting content of the request to determine a message type associated with the request;
• determining whether the message type indicates that the request is associated with dependent messages;
• determining a first number of concurrent sessions currently connected to the network device;
• identifying a second number of sessions associated with a regression point;
• determining a ratio between a current load of the network device and a capacity of the network device;
• and
• in response to one or more of:
• (a) the ratio between the current load and the capacity exceeding the threshold, (b) the message type indicating that the request is associated with dependent messages, and (c) the first number approaching the second number, responding, by the network device,
• determining, by the network device, a wait time period based on one or more of (a) the ratio between the current load and the capacity, and (b) the first number of concurrent sessions;
• and
• responding, by the network device, to the network node with a message including the wait time period.