Re: [multipathtcp] (2) Regarding rate control at a subflow level

[Nagesh shamnur <[e-mail]>]

From: Madhan Raj Kanagarathinam [mailto:[e-mail]]
Sent: 20 May 2019 11:57
To: Nagesh shamnur <[e-mail]>
Cc: Alexander Frömmgen <[e-mail]>; Olivier Bonaventure <[e-mail]>; [e-mail]; Ashutosh prakash <[e-mail]>
Subject: RE:(2) [multipathtcp] Regarding rate control at a subflow level


Dear Nagesh,



So, while going through the discussion archives, could
only find that, the peer(server) can throttle the speed for the entire
connection by publishing a smaller receiver window rather than for a
particular subflow. I feel, it would be a good idea if the peers can
exchange this information using the control packets.



>> There are quite some research going on SBR and related areas. But if there has to options exchanged (or control), it has to be secured to protect from mangling. Another challenge is that estimating the bandwidth and frequent changing might also be overhead in wireless.

In my research work, to control the same, we throttle the UE RTT by introducing additional delay and controlling the path manager. There are few methods of ACK throttling as well which might help the same. So I would suggest indirect control instead of direct option exchanged.

            Madhan, Thanks for taking note of the problem. Using control channel was one of the alternative which I thought and I agree it might not be the best one. But the issue is genuine and needs to be addressed. Regarding the protecting the header from maligning, it is true for any other as well which is part of MPTCP right?

            Any indirect control IMO would still be a patchy solution rather than directly addressing the problem, right?

Regards,

Madhan





--------- Original Message ---------

Sender : Alexander Frömmgen <[e-mail]<mailto:[e-mail]>>

Date : 2019-05-19 00:47 (GMT+5:30)

Title : Re: [multipathtcp] Regarding rate control at a subflow level


Hi,

We experimented with different schedulers that implement related behaviour, e.g., to only use cellular data if a target throughput can't be provided otherwise and only transfer the fraction of data on the cellular path that is required to achieve the target throughput (which is effectively a kind of dynamic rate limiting). See https://progmp.net for further information.

Best,

Alexander
Am 17. Mai 2019 11:59:54 MESZ schrieb Olivier Bonaventure <[e-mail]<mailto:[e-mail]>>:

Dear Nagesh,

                 Greetings. In case of Mobile deployments of MPTCP,
though the data rates are getting cheaper, still it would be wise not to
run the cellular path to full limit but to throttle to a certain extent
considering cost in mind or if server wants to limit the client at
subflow level, then I couldn’t find the support for the same in the
specification.

We have developed several prototypes that include this capability in the
Linux kernel.

So, while going through the discussion archives, could
only find that, the peer(server) can throttle the speed for the entire
connection by publishing a smaller receiver window rather than for a
particular subflow. I feel, it would be a good idea if the peers can
exchange this information using the control packets.

We could imagine an MPTCP option that provides the maximum rate on a
per-subflow level, but I was wondering whether the use case is not to
limit the bandwidth on the smartphone at the link level (i.e. multiple
tcp connections or udp flows) and not at the subflow level. Could you
precise your use case for the subflow level ?

Thanks


Olivier

________________________________

multipathtcp mailing list
[e-mail]<mailto:[e-mail]>
https://www.ietf.org/mailman/listinfo/multipathtcp

_______________________________________________

multipathtcp mailing list

[e-mail]<mailto:[e-mail]>

https://www.ietf.org/mailman/listinfo/multipathtcp





[cid:image001.gif@01D50F1A.69955BA0]

[http://ext.samsung.net/mail/ext/v1/external/status/update?userid=madhan.raj&do=bWFpbElEPTIwMTkwNTIwMDYyNjM0ZXBjbXM1cDE1Nzg2MzJiZjhhNzkxZWVmZTk3MGJiNzY3OGU5NjYyNSZyZWNpcGllbnRBZGRyZXNzPW5hZ2VzaC5zaGFtbnVyQGh1YXdlaS5jb20_]