Unformatted text preview: ters are typically not on the high-speed
backbone links then there is no problem as classi cation at
moderate speeds is quite practical. 2.6 Architectural Considerations We have used the term ow without de ning what we mean.
This was intentional, as the CSFQ approach can be applied
to varying degrees of ow granularity; that is, what constitutes a ow is arbitrary as long as all packets in the ow
follow the same path within the core. In this paper, for convenience, a ow is implicitly de ned as a source-destination
pair, but one could easily assign fair rates to many other
granularities such as source-destination-ports. Moreover,
the unit of ow" can vary from island to island as long
as the rates are re-estimated when entering a new island.
Similarly, we have not been precise about the size of these
CSFQ islands. In one extreme, we could take each router
as an island and estimate rates at every router; this would
allow us to avoid the use of complicated per- ow scheduling
and dropping algorithms, but would require per- ow classication. Another possibility is that ISP's could extend their
island of CSFQ routers to the very edge of their network,
having their edge routers at the points where customer's
packets enter the ISP's network. Building on the previous
scenario, multiple ISP's could combine their islands so that
classi cation and estimation did not have to be performed
at ISP-ISP boundaries. The key obstacle here is one of trust
between ISPs. Having presented the basic CSFQ algorithm, we now return
to discuss a few aspects in more detail.
We have used exponential averaging to estimate the arrival rate in Eq. 3. However, instead of using a constant
exponential weight we used e,T=K where T is the interpacket arrival time and K is a constant. Our motivation
was that e,T=K more closely re ects a uid averaging process which is independent of the packetizing structure. More
speci cally, it can be shown that if a constant weight is used,
the estimated rate will be sensitive to...
View Full Document
- Fall '10
- Eugene Ng
- Scheduling algorithm, Round-robin scheduling, Scheduling algorithms, Ow, Fair queuing, ows