1

IPv6 was developed to solve some of the problems of IPv4, such as QoS. We know in IPV4 based on TOS values, a packet would be placed in a prioritized outgoing queue,or take a route with appropriate latency, throughput, or reliability. The IPV6 header has a structure that identifies the flow of packets (Flow Label Field) and thereby directs it to the router. So IPV6 and IPV4 both have fields for QoS. We know that QoS was almost ignored by routers in IPV4.How exactly IPV6 observe packets priority?How we are sure about it?

4
  • Have you looked at the IPv6 specifiication? How this would be handled would surely be described within it.
    – Ramhound
    May 1, 2013 at 11:19
  • Yes.I've searched a lot.RFC 2401: "IPv4 and IPv6 IPSec are functionally identical. Although there are some slight differences in implementation, the actual authentication and encryption functionality is the same - it is even specified in the same Internet Engineering Task Force request for comments document" link
    – Hanna
    May 1, 2013 at 13:28
  • If they are functionaly identical what exactly is your question? I also wouldn't say that QOS was ignored by routers in IPV4 considering nearly every consumer grade router supports it.
    – Ramhound
    May 1, 2013 at 15:11
  • As I said,it's clear that IPV6 was developed to solve some of the problems of IPv4,and one of them is QoS.So if they are functionaly identical,how IPV6 solve IPV4 QoS problems?
    – Hanna
    May 1, 2013 at 20:24

1 Answer 1

2

It's worth noting that QoS in IPv4 is usually done using DSCP. DSCP is sort of an overhaul of the old TOS field: it occupies the same bits of the IPv4 header, but the meanings are different and generally better aligned with the kinds of QoS policies people really want to have.

Using DSCP basically makes IPv4 and IPv6 QoS the same, because the DSCP field is defined the same way for IPv4 and IPv6 (in different positions in the header).

QoS is not almost ignored by routers in IPv4. QoS is almost ignored by routers on the Internet. That goes for both IPv4 and IPv6. It's not a technical limitation, it's a trust & policy limitation. The Internet is a public network. Basically, none of the ISPs who forward your packets for you are inclined to trust you to tell them what kind of special treatment you want for your packets. As the originator of the packets, your self-interest would be to simply mark almost every one of your packets with the DSCP value that will result in the packet being placed in the best possible priority queue while hoping that at least some other Internet users won't do the same.

ISPs can and do classify traffic according to criteria which they control. For example, they might route all Internet traffic as best effort, route all traffic coming from leased private network customers using something better, and treat VoIP traffic related to the ISP's telco arm's internal network best of all. They might do it by encapsulating these different kinds of traffic with different MPLS labels, or they might rewrite DSCP bits in packets. Again, this is all IP-version-agnostic.

You can likewise mark traffic on your own internal network and have your own routers respect these markings to classify traffic into different priority queues. But you can't expect anyone else to respect your markings unless you are paying them do to so.

6
  • To clarify what I mean about priority: I've read somewhere:" IPv6 provides a way for applications to request handling without delay throughout the WAN. The term often used to describe this is low latency. Streaming audio and video requires low latency through high priority. To prevent a break down in the scheme, the various application can share connection via priority level."link
    – Hanna
    May 1, 2013 at 20:22
  • *My last comment:As I said,it's clear that IPV6 was developed to solve some of the problems of IPv4,and one of them is QoS.So if they're functionaly identical,how IPV6 solve IPV4 QoS problems?
    – Hanna
    May 1, 2013 at 20:28
  • Your assumption is wrong. IPv6 does not solve IPv4 QoS problems.
    – Celada
    May 1, 2013 at 21:26
  • That's what I found in different resources...link
    – Hanna
    May 2, 2013 at 13:56
  • 1
    Celada thanks for sharing your experiences,I think it may differs from theory to practice...
    – Hanna
    May 3, 2013 at 20:20

Not the answer you're looking for? Browse other questions tagged .