IDEAL INDUSTRIES UniPRO MGig1, User Manual

Page: 19 / 68

UniPRO 152810
User Guide Page 19
Traffic Impairments
Transmission tests check for impairment to Ethernet traffic in the form of Errors, Latency and Jitter.
Frame errors are either Lost Frames (frames that were transmitted but were lost in the network)
or Corrupted frames (frames that were received but were found to contain errors).
Payload errors
are errors in the data part of the Ethernet frames.
Latency is network propagation delay.
Jitter is short term variations in network propagation delay.
To detect the above listed impairments, tests are run under specific, user defined conditions, e.g. with
different frame sizes and frame rates, and in the presence of varying levels of traffic loading. Some tests
determine the maximum performance that a network connection can achieve without exceeding pre-
defined limits on impairment levels.
Ethernet transmission testing methods
Various methods of Ethernet transmission testing have been developed over the years. Each
development brought additional sophistication and value to the knowledge about the performance of the
network being tested. More sophistication increased the test times. Some tests can run for days or
weeks. The most recent developments have sought to reduce test times without losing valuable
information about the network performance.
BERT
Bit Error Ratio Test (sometimes incorrectly referred to as “Bit Error Rate Test”) is a traditional test method
used for many types of telecommunications transmission system. In Ethernet environments, it focuses on
testing the frame payload, putting pre-defined patterns of test data into the frames and checking them bit-
by-bit at the receiver. Individual bit errors are counted and expressed as a ratio to the total data received,
and periods of time during which the error ratio exceeds pre-defined limits are recorded. Errors of
different types can be forced into the frames or the payload data in order to confirm that the error checking
mechanism is working correctly and to determine how the network responds to errored data.
SLA-Tick
Service Level Agreements (SLAs) are commonly used to define a contractual commitment to provide
specified minimum performance of network connections. They normally define minimum Information Rate
and maximum Latency and Jitter. For the SLA-Tick test, multiple services can be selected to occupy the
available bandwidth, each with its own fixed frame size and each having its own performance limits. This
test always uses Layer 4 framing, because it can propagate through all types of network and because the
reduced payload is not relevant.
RFC2544
The Internet Engineering Task Force defined this testing methodology which comprises the following sub-
tests, each of which can be included or omitted from the overall test sequence. (Note, however, that some
of the sub-tests rely on the results of others before they can be run.) A full RFC2544 test with all sub-tests
at all frame sizes can take a long time to complete, so the tester provides a choice of test profiles relating
to the time needed to complete the tests. Layer 4 framing is used.
Throughput This is a test of achievable Information Rate at different frame sizes. The maximum
theoretical Information Rate decreases with decreasing frame size, because the overhead becomes more
significant compared to the payload. This sub-test compares the actual performance against theoretical
and target limits.