IDEAL INDUSTRIES UniPRO MGig1, Руководство пользователя

Страница: 20 / 68

152810 UniPRO
Page 20 User Guide
Latency Latency increases with increasing frame size because frames have to be buffered in
network equipment prior to onward transmission and the bigger the frame the longer it takes before it can
be forwarded. The Latency sub-test measures latency against theoretical and target limit for different
frame sizes.
Jitter This should be independent of frame size, and the jitter sub-test is designed to confirm this
and to reveal any network impairments which might be introducing jitter dependent on frame size.
Frame Loss As frame rate increases, networks will start to drop frames. For each frame size, this
sub-test varies the frame rate until the point is found at which frames just start to be dropped.
Back-to-Back Each Ethernet frame must be separated from the next by a minimum Inter Frame
Gap (IFG). For each frame size, the Back-to Back sub-test assesses the ability of the network to handle
frames separated by the minimum IFG, measuring the longest burst of these frames that can be
successfully propagated without errors.
System Recovery When network elements are overloaded by excessive traffic, they can take some
time to recover to a condition in which they are ready to transmit traffic again. For each frame size, the
System recovery sub-test generates traffic at the rate found by the Throughput sub-test and measures the
recovery time of the network.
Y.1564 (NetSAM)
In an attempt to overcome some of the limitations of RFC2544, notably the long test times, the sequential
nature of the tests and the unrealistic single-service traffic method, the International Telecommunications
Union defined a new Ethernet service activation test methodology - Y.1564. The first phase in this method
is a Service Configuration Test, which evaluates the ability of the network to achieve the Committed
Information Rate (CIR) and Excess Information Rate (EIR) without error for each of a multitude of
services, one at a time. The Service Performance Test then measures the Frame Transfer Delay
(Latency), the Frame Delay Variation (Jitter), the Frame Loss Ratio and the Service Availability
simultaneously for each service at the CIR. By doing this, the NetSAM test evaluates all of the Key
Performance Indicators of the network at once over multiple services simultaneously, much more
accurately simulating real-world traffic than RFC2544. It also polices the ability of the network to control
and handle traffic which is marked in color to indicate its priority in relation to the CIR and EIR defined in
the SLA. Layer 4 framing is used.