Re: [Detnet] WG adoption poll: draft-varga-detnet-pof-03

Hi Xuesong,
Many thanks for the support of adoption and your comments.
See my comments below. 
Cheers
Bala'zs

-----Original Message-----
From: detnet <detnet-bounces@ietf.org> On Behalf Of Gengxuesong (Geng Xuesong)
Sent: Friday, September 9, 2022 12:28 PM
To: Lou Berger <lberger@labn.net>; DetNet WG <detnet@ietf.org>
Cc: DetNet Chairs <detnet-chairs@ietf.org>
Subject: Re: [Detnet] WG adoption poll: draft-varga-detnet-pof-03

Hi WG,

Support adoption with some comments which could be considered by the authors as follows. 

Best
Xuesong
--
section 1
" 
A possible implementation of PRF and PEF functions is described in  [IEEE8021CB] and the related YANG model is defined in [IEEEP8021CBcv].

In general, use of per packet replication and elimination functions may result in out-of-order delivery of packets, which may not be acceptable for some deterministic applications.  Correcting packet order is not a trivial task, therefore details of a Packet Ordering Function (POF) are specified herein.
"
[Xuesong] It may be more clear if adding some background here, for example why PRF and PEF are defined in IEEE and POF is defined in IETF.

<Balazs> Sure we may add some further clarification. Please note, that the PREOF functions are defined by IETF. Here we refer to IEEE standards only as a possible "algorithm / implementation variant".

" So far in packet networks, out-of-order delivery situations were
   handled at higher OSI layers at the end-points/hosts (e.g., in the
   TCP stack when packets are sent to application layer) and not within
   a network in nodes acting at the Layer-2 or Layer-3 OSI layers."
[Xuesong] I think packet in order is also requested in network. For example per packet load balancing will also cause packets out of order and reordering is requested. So does this mechanism also work for this scenario?

<Balazs> Text here just refers to some history / examples. Regarding your question: yes, the mechanism described in the draft can work also in per packet load balancing scenarios. 

Section 4.2
"Delay calculator: buffering time considers (i) the delay
       difference of paths used for forwarding the replicated packets
       and (ii) the bounded delay requirement of the given DetNet flow."
[Xuesong] Why is the delay caused by reordering not considered in the delay calculator here?

<Balazs> It is considered. The delay calculator determines the maximum delay a packet can sit in the POF buffer (i.e., POFMaxDelay) without breaking the e2e latency requirement.

Section 4.3
"The basic POF algorithm delays all out-of-order packets until all
   previous packet arrives or a given time (POFMaxDelay) elapses."
[Xuesong] Shall we consider to add the a parameter of MaxDisorderNum here? Because the buffer size is limited and the maximum number of streams that can be processed (POF node capability) will be affected by this parameter, for example: the maximum number of streams = BufferSize/(MaxDisorderNum*MaxPacketSize)

<Balazs> In general, MaxDisorderNum can be calculated based on the flow characteristics and the network topology. It is similar to calculate the size of the "History Window" for PEF. We may need some discussion to clarify e.g., the relationship of MaxDisorderNum and MaxMisordering. My view is that, BufferSize and MaxDisorderNum are rather design parameters and not POF configuration parameters, but I may missed your point.

Section 4.6
" The selection of the POF algorithm depends on the network scenario
   and the remaining delay budget of a flow."
[Xuesong] Could here add some simple description about the existing normal reordering mechanism and the relationship with the algorithm described in this document, for example Sliding window?

<Balazs> This may need some further discussion. What do You mean by "normal reordering mechanism"? 

> -----Original Message-----
> From: detnet [mailto:detnet-bounces@ietf.org] On Behalf Of Lou Berger
> Sent: Friday, August 26, 2022 8:30 PM
> To: DetNet WG <detnet@ietf.org>
> Cc: DetNet Chairs <detnet-chairs@ietf.org>
> Subject: [Detnet] WG adoption poll: draft-varga-detnet-pof-03
> 
> Hello,
> 
> This email begins a 2-week adoption poll for:
> https://datatracker.ietf.org/doc/draft-varga-detnet-pof/
> 
> Please voice your support or technical objections to adoption on the 
> list by the end of the day (any time zone) September 9.
> 
> Thank you,
> Lou (as Co-chair)
> 
> _______________________________________________
> detnet mailing list
> detnet@ietf.org
> https://www.ietf.org/mailman/listinfo/detnet
_______________________________________________
detnet mailing list
detnet@ietf.org
https://www.ietf.org/mailman/listinfo/detnet