Re: [tcmtf] A terminological question: "small-packet flows"

You are right, Dan:

We are considering the use of a "period" policy, which sends a multiplexed
packet at the end of each interval. However, this policy should be combined
with a "MTU size limit" policy, which triggers the packet as soon as a "next
to MTU" multiplexed packet is sent, even if the period has not expired.

In addition, if a single packet is to be multiplexed, it will be better to
send it in its native form. We don't need the tunnel in that case.

So let's suppose we have identified a "small packet flow" and in a certain
moment it generates a burst of big packets. In that case, TCMTF still works.
If we use the "period+MTU size limit" policy, then big packets will be sent
alone as soon as they arrive to the multiplexer.

So, as you say, when initially defining "small-packet flows", it would be
useful to explain that the flow does not have to always be small packets for
the lifetime of the flow. We can say that in a "small-packet flow" the vast
majority of the packets are small.

Thanks!

Jose

De: tcmtf-bounces@ietf.org [mailto:tcmtf-bounces@ietf.org] En nombre de Dan
Wing
Enviado el: miércoles, 12 de junio de 2013 17:34
Para: jsaldana@unizar.es
CC: tcmtf@ietf.org; Mirko Sužnjević
Asunto: Re: [tcmtf] A terminological question: "small-packet flows"

On Jun 12, 2013, at 6:10 AM, Jose Saldana <jsaldana@unizar.es> wrote:

Hi all.

Mirko and I are working on an improved version of the "TCMTF -
recommendations" document. Since TCMTF is not only suitable for real-time
services, but also for non real-time ones (M2M, flows of ACKs, instant
messaging), one possibility is using the term "small-packet flows".

I agree TCMTF is suitable for small packets.  

But by saying "small packet flow", it implies that the *entire flow* has to
be small packets for TCMTF to kick in.  But TCMTF is useful if the flow is a
mix of small and large packets, and useful if the flow began with lots of
large packets (TLS or SSH handshake followed by a file transfer, or an HTTP
GET (with a 4KB cookie) followed by a file transfer).

I like the term "small-packet flows" and it is concise, but when initially
defined it would be useful to explain the flow does not have to always be
small packets for the lifetime of the flow.

-d

The advantages are clear:

- It is more generic.

- It includes the characteristics of TCMTF-able packets:

- low payload-to-header ratio

- long-term flows

This term is also being used in some technical documents:
<http://www.huawei.com/ilink/en/download/HW_193034>
www.huawei.com/ilink/en/download/HW_193034.

What do you think? Any other proposals?

Jose

_______________________________________________
tcmtf mailing list
 <mailto:tcmtf@ietf.org> tcmtf@ietf.org
 <https://www.ietf.org/mailman/listinfo/tcmtf>
https://www.ietf.org/mailman/listinfo/tcmtf