Skip to main content

What is Classification and marking? What is IP Precedence and DSCP Values? What is AF and EF?

 Classification and Marking

Classification and marking are the processes of identifying the priority of each packet. This is the first step of QOS control and should be done near the source hosts. Classification is the process of identifying and categorising traffic into classes, typically based upon incoming interface IP precedence, DSCP Source or Destination address. Application classification is the most fundamental QOS building block. Without classification, all packets are treated the same.




 



The table lists the criteria of classification                               



 

Marking 

Marking is the QOS feature component that colours a packet so it can be identified and distinguished from other packets in QOS treatment. Commonly used markers:

link-layer: COS (ISL, 802.1q), MPLS EXP bits, frame relay 

network layer: DSCP, IP precedence 



 

Layer 2 and Layer 3 Header Marking 



Let's take a look at the Layer 3 IPv4 packet ToS Byte. 

 



 

In the beginning, the ToS byte is defined like this ☝.



  • The starting 3 bits are used to define IP precedence. The higher the value means the higher the priority of IP Packets.
  • The rest of the bits are called the Type of Service. ToS bits are used to assign delay, throughput, and reliability.

Let's see the picture for better understanding.

 



 

 

But the “type of service” bits, which are a specified delay, throughput, and reliability, have never really been used. Only the IP precedence bits are used to assign a priority to the IP packets. Do not be confused by the name Tos 0-7 is called ToS byte, and 4-7 bits is called ToS bits (Type of Service). Look at the picture below👇. However, ToS bits (Type of Service) have never been used; only the IP Precedence is in use for prioritising the IP packets. 



Let's talk about what we are actually using nowadays?

The DS field (Differentiated Services), now we call the ToS byte the DS field, the name has changed.



DSCP (DiffServ Code Point) provides a 6-bit field for QoS marking in which the colour code (CS) is also called DSCP value. CS or DSCP  6 bits, among which 3 bits are the same as IP precedence, and the other 3 bits are ToS fields. Thus, the DSCP value range is 0 to 63. The graph below shows the DSCP and IP precedence bits:



DSCP value is also known as Per-Hop Behaviour PHB is Packets that are marked with a certain codepoint will receive a certain QoS treatment (for example, queuing, policing, or shaping). The default Per-Hop Behaviour PHB means that we have a packet that is marked with a DSCP value of 000000. This packet should be treated as “best-effort”.

 .

Now there are three types of defined PHBs:

Best-Effort (BE or DSCP 0) is the packet that is marked with a DSCP value of 000000. This packet should be treated as “best-effort”.

 Assured Forwarding (AF) Assured forwarding allows you to provide assurance of delivery as long as the traffic does not exceed some subscribed rate.

Expedited Forwarding (EF) is that any traffic class with EF's related DSCP is given the highest priority

 

let's see CS, AF, EF, and Best effort.



 

 

 

 The first bits are called CS, a higher number means high priority (same as IP Precedence), and the rest of the 3 bits are called AF higher number means high drop probability assured forwarding bits.



 



What are CS  and AF?

CS 1    +     AF 1     =  001 0100

CS 2   +     AF 2     =   010 1000

CS 3   +     AF 3     =   011 0110

 

How does this work?

Let's take an example, we have FastEthernet 0/0, and the queue is full, and the router wants to drop packets.

 


Comments

Popular posts from this blog

What is of NBAR (network- based application recognition),classification QOS How to configure NBAR?

 When we implement Quality of Service (QoS ) the first step is classification, by default our router does not care about what kind of IP packets it is. Our router just looks at the routing table to find the correct destination IP address and forwards. When we configure QoS technologies like queuing, policing, or shaping before doing first we have to identify what kind of traffic is running on our routers so we can figure out what kind of application it belongs to. This is classification. What is Quality of Service (QOS)? What is Modular Quality of service CLI (MQC)? What is Classification and marking? What are IP Precedence and DSCP Values? What is AF and EF?     Once we identify our traffic (classify). Now we can mark and apply QoS policy to it. There are a few methods on IOS routers for classification: Header inspection It is a simple classification method and it works really good but has some disadvantages. Let’s say we configure our router all th...

What are BGP RR Cluster and BGP RR refelector? CCIE LAB

  Route Reflector Cluster A BGP-enabled router or BGP speakers. By default BGP speaker routers do not advertise iBGP-learned prefixes to iBGP peer routers - this has to be done to maintain loop prevention. RFC4456 introduces the route reflection feature which helps to remove the need of full mesh configuration between iBGP speakers routers. When the route reflector reflects a route, it creates and modifies an optional non-transitive attribute called CLUSTER_LIST by adding its own cluster ID to it. This attribute is used for loop prevention: when a BGP router receives update which CLUSTER_LIST contains router's own cluster ID, this update is discarded. Route reflector Route Reflector scalable alternative to an ibgp full mesh its allow a  BGP speaker (route reflector –RR) to advertise route received from an ibgp peer to other ibgp peers. note:client update server. Server update to all the remaining clients. All client should establish neighbor with only servers . client will...