In this post, we will try to clarify key concepts around Netflow technology and potentially correct some common misconceptions. Particularly we will explain why the “what is the sampling-rate you support?” is not the right question.
We will describe in extensive details the NCS5500 implementation too.
Also, we will share what we measured in different networks in Europe and North America. This information will be helpful to understand the parameters of this equation.
We will provide tools to answer questions like:
- how many new flows per second?
- how long the flows exist and how long a representation of them will stay in cache?
- are we dropping samples because of the protection shaper?
