BLUE
: A New Class of Active Queue Management Algorithms
Wu-chang Feng
y
Dilip D. Kandlur
z
Debanjan Saha
z
Kang G. Shin
y
y
Department of EECS
z
Network Systems Department
University of Michigan IBM T.J. Watson Research Center
Ann Arbor, MI 48105 Yorktown Heights, NY 10598
Phone: (313) 763-5363 Fax: (313) 763-4617 Phone: (914) 784-7194 Fax: (914) 784-6205
f
wuchang,kgshin
g
@eecs.umich.edu
f
kandlur,debanjan
g
@watson.ibm.com
Abstract
In order to stem the increasing packet loss rates caused by an exponential increase in network traffic,
the
IETF
is considering the deployment of active queue management techniques such as R
ED
[13]. While
active queue management can potentially reduce packet loss rates in the Internet, this paper shows that
current techniques are ineffective in preventing high loss rates. The inherent problem with these queue
management algorithms is that they all use queue lengths as the indicator of the severity of congestion.
In light of this observation, a fundamentally different active queue management algorithm called B
LUE
is proposed. B
LUE
uses packet loss and link idle events to manage congestion. Using simulation and
controlled experiments, B
LUE
is shown to perform significantly better than R
ED
both in terms of packet
loss rates and buffer size requirements in the network. As an extension to B
LUE
, a novel technique for
enforcing fairness among a large number of flows is described. In particular, this paper proposes and
evaluates Stochastic Fair B
LUE
(
SFB
), a queue management algorithm which can identify and rate-limit
non-responsive flows using a very small amount of state information.