190
12.7 RFC 2544 Conformance Testing
12.7.1 Fundamentals
The RFC 2544 conformance testing was introduced as a method to benchmark interconnected network devices. Because
of its ability to measure throughput, burstability, frame loss and latency, this methodology is also used to test Ethernet-
based networks and is now the de facto standard when benchmarking an Ethernet network. The test methodology
defines the different frame sizes to be tested (64,128, 256, 512, 1024, 1280 and 1518 bytes), the test time for each test
iteration (typically at least 60 or 120 seconds and the frame format (IP/UDP), etc.
The throughput test allows the technician to obtain the maximum rate at which none of the offered frames are dropped
by the device/system under test (DUT/SUT). This measurement translates the obtained rate into the available
bandwidth of the Ethernet virtual connection.
The latency test (for store-and-forward devices) refers to the time interval that begins when the last bit of the input
frame reaches the input port and ends when the first bit of the output frame is seen on the output port. It is the time
taken by a bit to go through the network and back. Latency variability can be a problem. With protocols like VoIP and
IPTV, a variable or long latency can cause degradation in voice and video quality.
The frame loss test calculates the percentage of frames that should have been forwarded by a network device under
steady state (constant) loads that were not forwarded due to lack of resources. This measurement can be used for
To Next Page
Loading...