Re: [Apn] [DMM] Regarding APN Usecase in Mobile Core

Hi Sridhar, Xuesong,

Quick thoughts below [Uma]:

--
Uma C.

On Sun, Jan 31, 2021 at 8:42 PM Sridhar Bhaskaran <
sridhar.bhaskaran@gmail.com> wrote:

> Hi Gengxuesong,
>
> On,
>
> >> Maybe the existing question becomes whether DSCP is sufficient for
> passing information from the mobile network to transport network? whether
> there is any simplification or omission of information in the mapping to
> the DSCP?
>

[Uma]: First of all, a lot of discussion below was discussed during 04
version of https://tools.ietf.org/html/draft-clt-dmm-tn-aware-mobility-08.
It was presented by John in one the IETF DMM sessions that time.  And yes,
it was limiting and moreover a lot of implementations reset DSCP bits in
the shared transport scenarios (like WAN  scenario).

>
>
> 3GPP allows mapping of a GTP-u tunnel (PDU session) to an index to a
> transport network configuration / policy via "Network Instance" concept.
> When the GTP-u tunnel is setup in the UPF by the signalling plane, it
> provides a "Network Instance" which is just a string that acts as an index
> into UPF's transport layer configuration.
>

[Uma]:  Right and that's natural as overlay is processed at the tunnel
endpoints (gNB/PSA-UPF/ULCL-UPF/BP-UPF).

> Now what the transport layer configuration is - 3GPP doesn't care as it is
> out of their remit. If you are using just plain IP routing, then DSCP (for
> IPv4) and Flow label (IPv6) are the options to map a QFI within a GTP-u
> tunnel to IP transport. However if you are using - say a source routing
> technology like SRv6, you may potentially configure the UPF to map a slice
> (PDU session / TEID) and QFI to a specific SID list. If you use PPR
> (Preferred path routing), then MTCN-ID is an option.
>

[Uma]: I am afraid this may not be possible in all scenarios for example
gNB may not be knowing the all underlying transport information to encode
that way. However, if 5G-AN node is integrated with the PE this is a
possibility. Another possibility is SRv6 user plane where interoperability
scenarios  and overall user plane are described in
https://tools.ietf.org/html/draft-ietf-dmm-srv6-mobile-uplane-09

>
> Also note that on the N3 interface between UPF and gNB, if it is going to
> traverse a public network, operators would employ IPSec tunneling for
> security. So any markings at the transport protocol layer (e.g. UDP source
> port) will not be visible to on path routers due to IPSec. In such a case,
> the mapping of Slice (PDU Session / TEID) and QFI should be to the outer IP
> header in the IPSec tunnel. This is where a solution at IP header level
> helps (e.g DSCP / flow label / SID list)
>

[Uma]: Or one can use
https://tools.ietf.org/html/draft-xu-ipsecme-esp-in-udp-lb-07 (note RFC
3948 UDP encap only for nat traversal scenarios)

>
> Regards
> Sridhar
>
> On Mon, Feb 1, 2021 at 8:10 AM Gengxuesong (Geng Xuesong) <
> gengxuesong@huawei.com> wrote:
>
>> Hi Sridhar,
>>
>>
>>
>> Thanks for your explanation. It is much clearer for me as the following:
>>
>> ž   UPF < ----- GTP-u tunnel [PDU session]------ > gNB
>>
>> ž   PDU session contains multiple QoS flows [QFI marker->5QI]
>>
>> ž   QFI marker in GTP-u extension header
>>
>> 5QI is not in GTP-u header directly but could be indicated by QFI. So the
>> QoS requirements presented by 5QI can’t be used/seen by transport network
>> in the existing in the current implementation (TEID is similar). And DSCP
>> can act as a mediator between transport network and UPF/gNB.
>>
>> Maybe the existing question becomes whether DSCP is sufficient for
>> passing information from the mobile network to transport network? whether
>> there is any simplification or omission of information in the mapping to
>> the DSCP?
>>
>>
>>
>> Best
>>
>> Xuesong
>>
>>
>>
>>
>>
>> *From:* Apn [mailto:apn-bounces@ietf.org] *On Behalf Of *Sridhar
>> Bhaskaran
>> *Sent:* Friday, January 29, 2021 8:06 PM
>> *To:* Gengxuesong (Geng Xuesong) <gengxuesong@huawei.com>
>> *Cc:* apn@ietf.org; Uma Chunduri <umac.ietf@gmail.com>; Kaippallimalil
>> John <john.kaippallimalil@futurewei.com>; dmm <dmm@ietf.org>; Lizhenbin <
>> lizhenbin@huawei.com>
>> *Subject:* Re: [Apn] [DMM] Regarding APN Usecase in Mobile Core
>>
>>
>>
>> Hi Gengxuesong,
>>
>>
>>
>> QFI (mapped to 5QI) and TEID are to identify QoS flows and PDU sessions
>> within UPF and gNB. They are not meant for transport network to look into.
>> UPF and gNB map QFI to DSCP marking in the outer IP header which the
>> transport network can use for differentiated services.
>>
>>
>>
>> Regards
>>
>> Sridhar
>>
>>
>>
>> On Fri, Jan 29, 2021 at 3:37 PM Gengxuesong (Geng Xuesong) <
>> gengxuesong@huawei.com> wrote:
>>
>> Hi Sridhar,
>>
>>
>>
>> Thank you for your additional information. Can I come to a conclusion
>> that: both 5QI and TEID have the potential to provide additional
>> information for differentiated services in transport network, although the
>> two parameters act in different scopes?
>>
>>
>>
>> Best
>>
>> Xuesong
>>
>>
>>
>> *From:* Sridhar Bhaskaran [mailto:sridhar.bhaskaran@gmail.com]
>> *Sent:* Friday, January 22, 2021 12:39 PM
>> *To:* Kaippallimalil John <john.kaippallimalil@futurewei.com>
>> *Cc:* Gengxuesong (Geng Xuesong) <gengxuesong@huawei.com>; Uma Chunduri <
>> umac.ietf@gmail.com>; Lizhenbin <lizhenbin@huawei.com>; apn@ietf.org;
>> dmm <dmm@ietf.org>
>> *Subject:* Re: [DMM] [Apn] Regarding APN Usecase in Mobile Core
>>
>>
>>
>> Hi Xuesong,
>>
>>
>>
>> In addition to what John said, in 3GPP networks there is one GTP-u tunnel
>> per bearer (in case of 4G) and one GTP-u tunnel per PDU session (in case of
>> 5G).
>>
>>
>>
>> One UE (user equipment - i.e mobile device) may have multiple PDU
>> sessions and hence in the network there may be more than one tunnel for a
>> UE. The scope of GTP-u tunnel is from UPF to gNB only. GTP-u does not go
>> all the way upto UE. The GTP-u header has a field called "TEID" (tunnel
>> endpoint identifier). The TEID in the header identifies a context in UPF
>> and gNB. The context gets established through signalling plane. The context
>> provides information on the QoS to be provided for the bearer,  PDCP
>> ciphering keys applicable for the bearer context etc.,. If there are a
>> million UE that are getting connected to a UPF, there could be few million
>> GTP-u tunnels (TEID).
>>
>>
>>
>> In summary:
>>
>> 1. The 5QI / QFI marking in the GTP-u extension header provides a lookup
>> for the general QoS characteristic applicable for that 5QI
>>
>> 2. TEID in the GTP-u header provides a lookup for UE and bearer specific
>> contextual information for any differentiated treatment.
>>
>>
>>
>> Regards
>>
>> Sridhar
>>
>>
>>
>> On Fri, Jan 22, 2021 at 2:55 AM Kaippallimalil John <
>> john.kaippallimalil@futurewei.com> wrote:
>>
>> Hi Xuesong,
>>
>>
>>
>> Traffic policy for subscribers is managed per PDU session at the UPF (and
>> gNB).
>>
>> GTP-u does provide encapsulation between the end points, but its control
>> fields are meant for conveying control semantics between the GTP endpoints:
>> they were not intended for IP transport/ traffic underlays. 5QI/QCI etc are
>> in the GTP extension header which may not be ideal to lookup to classify
>> each packet in the transport network.
>>
>>
>>
>> The entity that classifies data packets (upstream at gNB and downstream
>> at UPF-PSA) also inserts the DSCP for that GTP packet. The classification
>> is based on subscriber aspects but may also on be based on its content
>> (e.g., using DPI).
>>
>>
>>
>> Best Regards,
>>
>> John
>>
>>
>>
>>
>>
>> *From:* dmm <dmm-bounces@ietf.org> *On Behalf Of *Gengxuesong (Geng
>> Xuesong)
>> *Sent:* Wednesday, January 20, 2021 8:23 PM
>> *To:* Uma Chunduri <umac.ietf@gmail.com>; Lizhenbin <lizhenbin@huawei.com
>> >
>> *Cc:* apn@ietf.org; dmm <dmm@ietf.org>
>> *Subject:* Re: [DMM] [Apn] Regarding APN Usecase in Mobile Core
>>
>>
>>
>> Hi Uma and all,
>>
>>
>>
>> I have read the document and got a few questions:
>>
>> In my understanding, in the UPF where traffic policy is enforced, the
>> fine-granularity services are provided. Then what fields in the GTP-u
>> encapsulation indicates the traffic's service requirements? When a GTP-u
>> tunnel goes into a SRv6 policy, according to which fields in the GTP-u
>> encapsulation the DSCP is generated? We know that there are parameters such
>> as 5QI/QCI and QFI, whether they are associated with a GTP-u tunnel?
>>
>>
>>
>> Best
>>
>> Xuesong
>>
>> *From:* Apn [mailto:apn-bounces@ietf.org <apn-bounces@ietf.org>] *On
>> Behalf Of *Uma Chunduri
>> *Sent:* Tuesday, January 19, 2021 3:17 AM
>> *To:* Lizhenbin <lizhenbin@huawei.com>
>> *Cc:* apn@ietf.org; dmm <dmm@ietf.org>
>> *Subject:* Re: [Apn] [DMM] Regarding APN Usecase in Mobile Core
>>
>>
>>
>> Hi Robin,
>>
>>
>>
>> In-line..
>>
>>
>>
>> Cheers!
>> --
>>
>> Uma C.
>>
>>
>>
>> On Mon, Jan 18, 2021 at 5:25 AM Lizhenbin <lizhenbin@huawei.com> wrote:
>>
>> Hi APNers and DMMers,
>>
>> I remember that in the mobile core scenarios the GTP-u tunnel can be set
>> up according to the user and application requirements, but I do not
>> understand the details.
>>
>>
>>
>> [Uma]: Obviously, the best reference for GTP-U is TS 29.281. However, uou
>> should look into
>> https://datatracker.ietf.org/doc/draft-ietf-dmm-5g-uplane-analysis/
>> <https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdatatracker.ietf.org%2Fdoc%2Fdraft-ietf-dmm-5g-uplane-analysis%2F&data=04%7C01%7Cjohn.kaippallimalil%40futurewei.com%7C77632db844b34a8e561108d8bdb38cb8%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637467926158632424%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000&sdata=5nVGUJhkHBn7FWHUCFPC4lGhwr6lBqtIWNyY1o0dxsg%3D&reserved=0>
>> where lot more details and other references related this topic was analyzed
>> (primarily started after/during REL-15, when  any other use plane other
>> than GTP-U is worthwhile is debated for 5G N9 interface).
>>
>>
>>
>> I think when the packet tunneled by GTP-u traverses the APN-based
>> transport network, it may be mapped to the corresponding tunnel according
>> to the user and application requirements to implement the uniform service.
>> If you are familiar with the principle of GTP-u in the mobile core, please
>> help provide some details.
>>
>>
>>
>>
>>
>> Best Regards,
>>
>> Zhenbin (Robin)
>>
>>
>>
>>
>>
>> _______________________________________________
>> dmm mailing list
>> dmm@ietf.org
>> https://www.ietf.org/mailman/listinfo/dmm
>> <https://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.ietf.org%2Fmailman%2Flistinfo%2Fdmm&data=04%7C01%7Cjohn.kaippallimalil%40futurewei.com%7C77632db844b34a8e561108d8bdb38cb8%7C0fee8ff2a3b240189c753a1d5591fedc%7C1%7C0%7C637467926158632424%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C2000&sdata=oHJnGPO3hUL9xPkObVLsJ5S1JnwGOF6IeJrLADLrTVA%3D&reserved=0>
>>
>> _______________________________________________
>> dmm mailing list
>> dmm@ietf.org
>> https://www.ietf.org/mailman/listinfo/dmm
>>
>>
>>
>> --
>>
>>  o__
>>  _>  /__
>> (_) \(_)... Burn fat not fuel - Bike along to a healthier life and
>> cleaner
>> world! :)
>>
>> Sridhar Bhaskaran
>>
>>
>>
>> --
>>
>>  o__
>>  _>  /__
>> (_) \(_)... Burn fat not fuel - Bike along to a healthier life and
>> cleaner
>> world! :)
>>
>> Sridhar Bhaskaran
>>
>
>
> --
>  o__
>  _>  /__
> (_) \(_)... Burn fat not fuel - Bike along to a healthier life and cleaner
> world! :)
>
> Sridhar Bhaskaran
>
>