Traffic Management for the Available Bit Rate (ABR) Service in ...
Traffic Management for the Available Bit Rate (ABR) Service in ... Traffic Management for the Available Bit Rate (ABR) Service in ...
The scheme requires no parameter settings. Performance analysis with xed and variable ABR capacity. The drawbacks of the scheme are: The scheme does not measure the load (aggregate input rate) at the switch. As a result, if a source is sending at a rate below its CCR, then the bottleneck will be underutilized. The scheme also does not observe the queuing delay at the switch. Errors in estimation of ABR capacity result in errors in feedback andeventually result in queues. Hence, there is a possibility of in nite queues if the queuing delay is not considered as a metric. However, such a mechanism may easily be developed on similar lines as the ERICA+ proposal studied later in the dissertation. The scheme assumes that the sum of the number of bottlenecked and satis ed connections is equal to the number of connections setup. The scheme does not measure the number of active connections. As a result, if a connection is setup, but remains idle for a while, the allocations to other connections remain low and may result in underutilization. The convergence time is slower since the scheme attempts never to over-allocate (conservative). This non-optimistic strategy may result in link underutilization of the sources are not always active, or cannot utilize their ER allocations [2]. 79
4.10 DMRCA scheme The Dynamic Max Rate Control Algorithm (DMRCA) scheme [20] was developed by Chiussi, Xia and Kumar at Lucent Technologies, in an attempt to improve the EPRCA scheme. 4.10.1 Key Techniques DMRCA uses a rate marking threshold similar in concept to the MACR of EPRCA. However, the DMRCA threshold is a function of the degree ofcongestion at the switch and the maximum rate of all active connections. This rate threshold is used to esti- mate the maximum fairshare of any active connection on the link. The authors observe that the EPRCA depends upon the mean cell rate of all connections which it uses as a rate marking threshold. If this mean is close to the fairshare of available bandwidth on the link, then EPRCA performs well. But, if the approximation does not hold, then EPRCA introduces considerable unfairness. For example, if some connections are bottlenecked in other switches, they may cause underestimation of the fairshare. Another case is when rates oscillate due to transient behaviors and/or interactions with multiple switches, leading to incorrect estimates of the actual rate of the connection. The authors propose to use the maximum rate of all the active connections in- stead of the mean rate used by EPRCA. They observe that the maximum rate of all connections quickly rises to be above the desired \fairshare" (the maximum rate allocation for unconstrained connections at this switch). Further, this value can be made to converge to fairshare in the steady state. 80
- Page 55 and 56: It has been incorrectly believed th
- Page 57 and 58: NI CI Action 0 0 ACR Min(ER, ACR +
- Page 59 and 60: Source Rule 13: Sources can optiona
- Page 61 and 62: Switch Rule 1: This rule speci es t
- Page 63 and 64: Observe that the ABR tra c manageme
- Page 65 and 66: complex method like per-VC queuing
- Page 67 and 68: graphs have a steady state with con
- Page 69 and 70: Figure 3.2: Operating point between
- Page 71 and 72: The following example illustrates t
- Page 73 and 74: mention this concept of fairness fo
- Page 75 and 76: control problem was also simpler. S
- Page 77 and 78: of such an occurrence is expected t
- Page 79 and 80: y giving only one feedback per meas
- Page 81 and 82: CHAPTER 4 SURVEY OF ATM SWITCH CONG
- Page 83 and 84: The adaptive FCVC algorithm [63] co
- Page 85 and 86: np. Thus, long path VCs have fewer
- Page 87 and 88: networks like ATM can have complete
- Page 89 and 90: scheme followed by a discussion of
- Page 91 and 92: exibility of decoupling the enforce
- Page 93 and 94: 4.6.1 Key Techniques In EPRCA, the
- Page 95 and 96: If the mean ACR is not a good estim
- Page 97 and 98: is the ratio of the input rate to t
- Page 99 and 100: which shares the link equally as al
- Page 101 and 102: proportional to the the unused ABR
- Page 103 and 104: The key technique in the scheme is
- Page 105: the bandwidth allocations to satis
- Page 109 and 110: on its CCR. Further MAX times out i
- Page 111 and 112: other words, a di erent set of para
- Page 113 and 114: A combination of several ideas in a
- Page 115 and 116: 4.13 SP-EPRCA scheme The SP-EPRCA s
- Page 117 and 118: RTD and shuts o the source (for sta
- Page 119 and 120: 4.14.1 Common Drawbacks Though the
- Page 121 and 122: The ATM Tra c Management standard a
- Page 123 and 124: 5.1.1 Control-Cell Format The contr
- Page 125 and 126: The last two elds are used in the b
- Page 127 and 128: Figure 5.3: Flow chart for updating
- Page 129 and 130: LAF in cell Max(LAF in cell, z) The
- Page 131 and 132: the time of departure (instant mark
- Page 133 and 134: the steady state. The system operat
- Page 135 and 136: 5.2.3 Use Measured Rather Than Decl
- Page 137 and 138: said about the maximum queue length
- Page 139 and 140: The key problem with some unipolar
- Page 141 and 142: Figure 5.6: Space time diagram show
- Page 143 and 144: As noted, these heuristics do not g
- Page 145 and 146: Transmitted Cell Rate ABR Queue Len
- Page 147 and 148: 4. The RM cell contains a timestamp
- Page 149 and 150: 1. The source o ered average cell r
- Page 151 and 152: Transmitted Cell Rate Link Utilizat
- Page 153 and 154: Transmitted Cell Rate Link Utilizat
- Page 155 and 156: Transmitted Cell Rate Link Utilizat
4.10 DMRCA scheme<br />
The Dynamic Max <strong>Rate</strong> Control Algorithm (DMRCA) scheme [20] was developed<br />
by Chiussi, Xia and Kumar at Lucent Technologies, <strong>in</strong> an attempt to improve <strong>the</strong><br />
EPRCA scheme.<br />
4.10.1 Key Techniques<br />
DMRCA uses a rate mark<strong>in</strong>g threshold similar <strong>in</strong> concept to <strong>the</strong> MACR of EPRCA.<br />
However, <strong>the</strong> DMRCA threshold is a function of <strong>the</strong> degree ofcongestion at <strong>the</strong> switch<br />
and <strong>the</strong> maximum rate of all active connections. This rate threshold is used to esti-<br />
mate <strong>the</strong> maximum fairshare of any active connection on <strong>the</strong> l<strong>in</strong>k.<br />
The authors observe that <strong>the</strong> EPRCA depends upon <strong>the</strong> mean cell rate of all<br />
connections which it uses as a rate mark<strong>in</strong>g threshold. If this mean is close to <strong>the</strong><br />
fairshare of available bandwidth on <strong>the</strong> l<strong>in</strong>k, <strong>the</strong>n EPRCA per<strong>for</strong>ms well. But, if<br />
<strong>the</strong> approximation does not hold, <strong>the</strong>n EPRCA <strong>in</strong>troduces considerable unfairness.<br />
For example, if some connections are bottlenecked <strong>in</strong> o<strong>the</strong>r switches, <strong>the</strong>y may cause<br />
underestimation of <strong>the</strong> fairshare. Ano<strong>the</strong>r case is when rates oscillate due to transient<br />
behaviors and/or <strong>in</strong>teractions with multiple switches, lead<strong>in</strong>g to <strong>in</strong>correct estimates<br />
of <strong>the</strong> actual rate of <strong>the</strong> connection.<br />
The authors propose to use <strong>the</strong> maximum rate of all <strong>the</strong> active connections <strong>in</strong>-<br />
stead of <strong>the</strong> mean rate used by EPRCA. They observe that <strong>the</strong> maximum rate of<br />
all connections quickly rises to be above <strong>the</strong> desired \fairshare" (<strong>the</strong> maximum rate<br />
allocation <strong>for</strong> unconstra<strong>in</strong>ed connections at this switch). Fur<strong>the</strong>r, this value can be<br />
made to converge to fairshare <strong>in</strong> <strong>the</strong> steady state.<br />
80