ietf-wg-ccwg / rfc5033bis

IETF drafts
Other
4 stars 6 forks source link

Flows sharing a bottleneck under Single Algorithm Behavior? #86

Closed renghardt closed 8 months ago

renghardt commented 8 months ago

This might be an editorial issue, and/or I might be misunderstanding something:

Section 3.1 is explicitly scoped as Single Algorithm Behavior.

Subsection 3.1.4 starts with:

 In contexts where differing congestion control algorithms are used, it is important to understand whether an alternate congestion control algorithm can induce more harm to sharing flows than existing defined methods.

Doesn't the use of "differing congestion control algorithms" imply that this subsection is about Mixed Algorithm Behavior, so it should be under Section 3.2? Or am I missing a subtle nuance here, in which case I suggest clarifying why this text needs to be under 3.1? Or does it apply to both Single and Mixed Algorithm Behavior?

gorryfair commented 8 months ago

I changed to use the word "differing" ... was that the correct intent?

renghardt commented 8 months ago

Sorry, I don't follow, Gorry. My question was: Why is text about "differing algorithms" under Section 3.1 "Single Algorithm Behavior"? And should this text be under Section 3.2 "Mixed Algorithm Behavior" instead?

martinduke commented 8 months ago

I agree with Reese. This subsection is redundant with the reference to bottlenecks at the start of the section, and further discussion of "significant negative impact" in the mixed algorithm section.

martinduke commented 8 months ago

Fixed by #89