. . . . .
Low Latency Queuing
355
COMPUTING MAXIMAL DELAY
The delay value defined for a class determines the number of packets that can
be queued in the Low Latency queue, before drops begin. When computing the
maximal delay of a Low Latency class, consider both the maximal delay that
streams matching the class can tolerate, and the minimal delay that Check Point
QoS can guarantee.
Use the following method to estimate the greatest delay the class can tolerate:
1. Refer to the technical details of the streaming application, and find the
delay it can tolerate.
2. Find or estimate the delay the external network imposes.
3. Subtract the delay imposed by the external network from the maximum
delay the streaming application can tolerate.
Use the following method to estimate the smallest delay for the class:
1. Find the bit rate of the streaming application.
If you cannot find the bit rate of the streaming application in the
application properties, you can use SmartView Monitor to
observe the application and determine the bit rate.
2. Estimate the typical packet size in the stream.
If you do not know the packet size, you can use
fw monitor
to
capture traffic and record packet size, or you can use the LAN’s
MTU (1500 bytes for Ethernet).
3. Estimate burst size, by monitoring the internal interfaces that precede the
QoS Module.
If no burstiness is detected, the minimum delay of the class should be no
smaller than this:
(3 * packet size) / bit rate
