Switches do mainly have two forwarding strategies available to them:
- receive a frame completely into the buffer, evaluate the destination address, send frame from buffer to destination
- receive the frame's header into the buffer, evaluate the destination address and make forwarding decision, start sending frame to destination as data comes in
The first is generally referred to as store-and-forward, the second as cut-through. As you have already noted, there may be many definitions for "latency" in each of these scenarios, but these two are mainly used and even found their way into RFC 1242 (section 3.8):
- cut-through latency is the time between the reception and sending of the first byte of a particular frame (First-in-first-out latency),
- store-and-forward latency is measured as Last-in-first-out latency i.e. the time between the reception of the last frame byte and the sending of the first frame byte
Although 300 ns looks sane for a Gigabit interface (the Ethernet header including the preamble is 24 bytes long which equals to 192 bits and would take 192 ns to receive at a rate of 10^9 bits per second, leaving 108 ns for the forwarding decision), Intel could have chosen a very own definition for its numbers - you would need to ask a sufficiently savvy representative for a definitive statement.
A recently published whitepaper from Juniper ("Printed on recycled paper"!) provides a nice summary and a comprehensive explanation of this topic.