R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....Also note that their models do not include the presence of ABR traffic which is controlled by other switches. Raj Jain has made the best known contributions to the field of ATM ABR congestion control. His implementation friendly Explicit Rate Indication for Congestion Avoidance (ERICA) algorithm [17], its predecessor, the ERPCA [19] and its successor, ERICA [20] work well in a large number of situations and appear to be favored by ATM switch designers. ERICA is computationally inexpensive to implement (as compared to the other contributions mentioned above) and has been shown, via ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A Sample Switch Algorithm," AF-TM 95-0178R1, February 1995.

....Also note that their models do not include the presence of ABR traffic which is controlled by other switches. Raj Jain has made the best know contributions to the field of ATM ABR congestion control. His implementation friendly explicit rate indication for congestion avoidance (ERICA) algorithm [10] and its successor, ERICA [12] work well in a large number of situations and appear to be favored by ATM switch designers. ERICA is computationally inexpensive to implement (as compared to the other contributions mentioned above) and has been shown, via simulations, to rapidly achieve max min ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A sample switch algorithm ", AF-TM 95-0178R1, Feb. 1995.

.... the rate based congestion control algorithm [1, 3] These include EPRCA (Enhanced Proportional Rate Control Algorithm) 4] CAPC (Congestion Avoidance using Proportional Rate Control) 5] APRC2 (Adaptive Proportional Rate Control) 6] and ERICA (Explicit Rate Indication for Congestion Avoidance) [7]. Each algorithm has its own advantages and disadvantages in terms of, for example, effectiveness, robustness, fairness and configuration simplicity. We first summarize a recently proposed switch algorithm called as the max min scheme [8] A strong point of this algorithm compared with others is ....

....that is, target utilization Theta BW ) of the bandwidth is shared by ABR connections, and the rest of the bandwidth is not allocated to absorb the rate fluctuation. The load averaging interval is an interval for monitoring the current traffic load at the switch. Readers should refer to [7] for details of ERICA. Each graph shows ACRs of source end systems and queue lengths of switches. As can be found from these figures, the queue length grows when the new connection is activated (around t = 20, 40 and 60 ms) and the maximum queue length is about 470 cells in the LAN environment. ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

....In this configuration, different simulations were performed for different values of 1 and 2 . Each TCP connection used one ATM connection. Two ATM service classes were tested: UBR service class, without any kind of selective dropping mechanism; and ABR, implemented with the ERICA algorithm [13]. In both cases, the buffer size at each 0 200 400 600 800 1000 1200 1400 1600 1800 2000 0 2 4 6 8 10 12 14 16 x 10 7 Time (ms. bps Figure 10: TCP transmitting long application messages over UBR: aggregated traffic at the trunk link (bit s) 0 200 400 600 800 1000 1200 1400 1600 ....

R. Viswanathan R. Jain, S. Kalyanaraman and R. Goyal, "A sample switch algorithm," Contribution 95-0178, ATM Forum, 1995.

....Fahmy at the Ohio State University and San qi Li and his student Cathy Fulton at the University of Texas, Austin. Raj Jain made the best know contributions to the field of ATM ABR congestion control. His implementation friendly Explicit Rate Indication for Congestion Avoidance (ERICA) algorithm [17], its predecessor, the ERPCA [19] and its successor, ERICA [20] work well in a large number of situations and appear to be favored by ATM switch designers. 19 The basic ERICA algorithm is characterized by calculating two rates for each VC and writing the larger rate into the explicit rate ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A Sample Switch Algorithm," AFTM 95-0178R1, February 1995.

.... the rate based congestion control algorithm [1, 2] These include EPRCA (Enhanced Proportional Rate Control Algorithm) 3] CAPC (Congestion Avoidance using Proportional Rate Control) 4] APRC2 (Adaptive Proportional Rate Control) 5] and ERICA (Explicit Rate Indication for Congestion Avoidance) [6]. Each algorithm has its own advantages and disadvantages in terms of, for example, effectiveness, robustness, fairness and configuration simplicity. We first summarize a recently proposed switch algorithm called as the max min scheme [7] A strong point of this algorithm compared with others is ....

....that is, target utilization Theta BW ) of the bandwidth is shared by ABR connections, and the rest of the bandwidth is not allocated to absorb the rate fluctuation. The load averaging interval is an interval for monitoring the current traffic load at the switch. Readers should refer to [6] for details of ERICA. Each graph shows ACRs of source end systems and queue lengths of switches. As can be found from these figures, the queue length grows when the new connection is activated (around t = 20, 40 and 60 ms) and the maximum queue length is about 470 cells in the LAN environment. ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

....algorithms that compute the j values for the links. This is an important research problem. A survey is provided in [Bonomi and Fendick 95] an important seminal algorithm is presented in [Charny 94] an algorithm that has seen a lot of real implementation and experimentation is reported in [Jain et al. 95] Our work, which addresses the important issues of the effects of propagation delays, and time varying available link capacities (owing to high priority stream traffic) is documented in [Abraham and Kumar 97a, Abraham and Kumar 98a, Abraham and Kumar 98b, Abraham 98] Our approach yields ....

R. Jain, S. Kalyanraman, R. Viswanathan and R. Goyal, " A Sample Switch Algorithm", ATM Forum/95-0178, February 1995.

....has been standardized for ABR (Available Bit Rate) service class by the ATM forum [1, 2, 3, 4] The target of the standard is an operation algorithm of both source and destination end systems. Although some example behaviors of intermediate switches are introduced in the standard activities [5, 6, 7], implementation issues regarding intermediate switches are left to manufactures. In this paper, we will focus on the simplest switch among them, which is referred to as an EFCI bit setting switch or a binary switch. In the binary switch, the congestion is detected by a predefined threshold in the ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

....Max Min fair allocation is based on the idea that no session should get extra bandwidth at the expense of a session that has lesser than or equal bandwidth. This notion of Max Min fair allocation was subsequently adopted for ABR traffic. Charny [9] L. Benmohammed et al. [5] Raj Jain et al. [14] have developed fair allocation algorithms for the zero MCR case based on this notion of fairness. However this notion does not incorporate a minimum rate requirement for the sessions. Max Min Fair Rate Control of ABR Connections S.P. Abraham 2 Our work, reported here, focuses on the fair ....

R. Jain, S. Kalyanraman, R. Viswanathan and R. Goyal, " A Sample Switch Algorithm ", ATM Forum/95-0178, February 1995.

....has been standardized for ABR (Available Bit Rate) service class by the ATM forum [1, 2, 3, 4] The target of the standard is an operation algorithm of both source and destination end systems. Although some example behaviors of intermediate switches are introduced in the standard activities [5, 6, 7], implementation issues regarding intermediate switches are left to manufactures. In this paper, we will focus on the simplest switch among them, which is referred to as an EFCI bit setting switch or a binary switch . In the binary switch, the congestion is detected by a predefined threshold in ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

....transmission with variable rate coding (Hayden [12] Mosely [19] Hayden s algorithm consisted of a simple additive successive approximation update at each node. Mosely [19] extended Hayden s algorithm to accomodate asynchronous updates and network delays. A similar class of algorithms (ERICA [14] ,UT [10] have been proposed in recent literature using multiplicative successive approximation updates at each node. In other recent literature [9] 15] 21] several authors have presented methods for implementing a centralised max min fair sharing algorithm [7] in a distributed manner. Another ....

R. Jain, S. Kalyanraman, R. Viswanathan and R. Goyal, " A Sample Switch Algorithm", ATM Forum/95-0178, February 1995.

....explicit rate MMF algorithms attempted basically to implement variations of the well known centralised algorithm (see [6] in a distributed fashion; the algorithms reported in [9] and [18] are important examples of this approach. A combination of clever heuristics gave rise to the ERICA algorithm [17], which was adopted almost as a benchmark by the ATM forum, and has seen many implementations. In our work, we have shown the MMF rate allocation problem as being equivalent to obtaining the root of a certain vector equation, and have then developed a provably convergent algorithm using the ....

R. Jain, S. Kalyanraman, R. Viswanathan and R. Goyal, " A Sample Switch Algorithm", ATM Forum/95-0178, February 1995.

....Hong Kong eetsang ee.ust.hk Abstract In this paper, a new rate based switch mechanism for ABR traffic in ATM networks, which aims to rapidly achieve max min fairness allocation, is proposed. Simulation results show that the proposed scheme can out perform both CAPC [AF94 0983] and ERICA [AF95 0178] in terms of response times and peak queue lengths. An analytical approximation of the performance is also introduced and its accuracy is found to be close to the simulation results. A variant of the proposed scheme is presented for handling the problem of different source to bottleneck ....

....than those which start early [Jain95] The scheme may also result in unnecessary oscillations [AF94 0882] 2. 2 Explicit Rate Indication for Congestion Avoidance (ERICA) To tackle the problem of using queue length as overload indicator, Explicit Rate Indication for Congestion Avoidance (ERICA) [AF95 0178] is proposed. The idea is to use the queue growth rate instead as the overload indicator. The switch measures the time T for N cell arrivals. If the available capacity of the link is C cells per second and the desired target utilization is U , the overload factor can be computed as follows: ....

[Article contains additional citation context not shown here]

R. Jain, S. Kalyanaraman, R Viswanathan and R. Goyal, "A Sample Switch Algorithm," ATM Forum 95-0178R1.

....upstream and the downstream paths, the current bottleneck of the VC can quickly be found. The bandwidth allocation for different VCs can then be adjusted by using this new bottleneck information to achieve max min fairness allocation [1] We compare the proposed scheme to CAPC [2] and ERICA [3]. Simulation results showed that the transient response times of the sources are significantly reduced in the proposed scheme. Furthermore, the peak queue lengths of the switches are generally smaller. 1 Introduction One of the challenges in developing Asynchronous Transfer Mode (ATM) is the ....

....length as overload indicator may lead to unfairness [8] The scheme may also result in unnecessary oscillations [9] 2. 2 Explicit Rate Indication for Congestion Avoidance (ERICA) Instead of using queue length as the overload indicator, Explicit Rate Indication for Congestion Avoidance (ERICA) [3] uses the queue growth rate as the overload indicator. The switch measures the time T for N cell arrivals. If the available capacity of the link is C cells per second and the target utilization is U , the overload factor can be computed as follows: Overload F actor = N= T U C) 4) At the end of ....

[Article contains additional citation context not shown here]

Jain, R.: A Sample Switch Algorithm. ATM Forum 95-0178R1

.... monitorOfl ronito and high computing power The per connection monitoron uses sever#parO8OflV settings toguar5 tee thecor4SF5#3F of themonitorS rnitor If there#FO did notrt#Ofl the actual networ status, it ishar for thefair rir to converVS ERICA (Explicit Rate Indicationfor Congestion Avoidance) [11] is a well known examplefor this class of scheme. The switching components ar mainly consisting of the ASICs (Application Specific Integr2SS Cir2SS# and the pr cessor An e#ective way to achieve high switching speed is using the ASIC to per245 most of the switching functions. Although the prcessor ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A sample switch algorithm," ATM Forum cont. 95-0178R,Reb 1995; availabf at http://www.cis.ohio-state.edu/ jain.

.... Max Min Scheme with Delayed Adjustment (MMDA) The proposed Max Min Scheme with Delayed Adjustment (MMDA) is based on the Max Min scheme [2] The Max Min scheme can achieve both the max min fairness within the shortest time and the shortest peak queue length when compared to CAPC [5] and ERICA [6]. Since sometimes it is necessary to carry the delay information to the sources, a new field in the RM cell, called delay is proposed. Similar to other time related parameters, it is 3 bytes long [3] During the call setup, every switch participates in calculating the round trip time by adding ....

Raj Jain, "A Sample Switch Algorithm," ATM Forum 95-0178R1.

....share for all active sources. Several ER switch algorithms were proposed in the literature, all based on different ways to measure the actual ABR traffic currently crossing the switch. CLASS presently implements the ERICA (Explicit Rate Indication for Congestion Avoidance) algorithm, proposed in [11, 12], as well as its new version, named ERICA [12] 4.3 The Wireless Access Loop The approaches for the provision of services to mobile users through ATM networks are far from being standardized; the definition of a wireless access model in CLASS was thus based on the research work currently ....

....that the blocking probabilities for CBR and VBR connections are close to 0.01, and mapping TCP connections on either a UBR service with shaping at 10 Mbit s, or an ABR service. In the latter case, ABR control is achieved using the ERICA (Explicit Rate Indication for Congestion Avoidance) algorithm [11] with target utilization 0.98. Cell level simulations were run for 5 different configurations generated by the call level simulations; these configurations were deemed to be the worst cases observed during several hours of simulated network activity. This analysis was restricted to the case fl 0 ....

R.Jain, S.Kalyanaraman, R.Viswanathan, R. Goyal, "A Sample Switch Algorithm ", ATM Forum-TM 95-0178R1, February 1995

.... (MMRCA) 5] dynamic max rate control algorithm (DMRCA) 6] and congestion avoidance using proportional control (CAPC) 7] Typical exact fair rate calculation algorithms are: congestion control with explicit rate indication (CCERI) 8] explicit rate indication for congestion avoidance (ERICA) [9], efficient rate allocation algorithm (ERAA) 10] and fast Max Min rate allocation algorithm (FMMRA) An excellent comparison of these schemes can be found in [2] The EPRCA uses several important ideas that were developed through the years [11, 12, 13, 14] The EPRCA with Intelligent Marking, ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A Sample Switch Algorithm" ATM Forum cont. 95-0178R, February 1995; http://www.cis.ohio-state.edu/ jain. 21

....that the blocking probabilities for CBR and VBR connections are close to 0.01, and mapping TCP connections on either a UBR service with shaping at 10 Mb s, or an ABR service. In the latter case, ABR control is achieved using the ERICA (Explicit Rate Indication for Congestion Avoidance) algorithm [9] with target utilization 0.98. The results that we obtained from cell level simulations are reported in Table 1. The four columns with numerical values refer to ABR and UBR, with either B = 10, or B = 100, and with fl 0 = 2. The first three rows provide goodput results in Mb s for CBR, VBR, and ....

R.Jain, S.Kalyanaraman, R.Viswanathan, R.Goyal, "A Sample Switch Algorithm", ATM Forum-TM 95-0178R1, February 1995 65/12

....more detailed description of the required computations see [4] 2. 2 The OSU Congestion Avoidance Scheme This scheme was developed by Jain et al. 5] at the Ohio State University (OSU) In this scheme, the switches measure their input rates over a fixed averaging interval that is counted in cells [6]. Thereby, the input rate is calculated as follows: Input rate = number of received cells length of measurement interval With this calculated input rate a load factor can be computed which can then be used to indicate the over or underload state of the switch. load factor = input rate Target ....

....This problem can be solved by using larger measurement intervals. However, this reduces the reactivity of the algorithm and leads to longer transient periods. Actually, in our tests we already used a measurement interval of 60 cells which is already twice as large as the suggested value in [6]. Connection VC 0 VC 1 VC 2 VC 3 VC 4 VC 5 VC 6 F 1 3.63 3.57 7.5 3.65 3.58 6.06 8.8 2 3.58 3.59 7.67 3.68 3.58 6.1 11.42 3 3.60 3.57 6.2 3.66 3.59 9.22 11.92 average 3.60 3.57 7.12 3.66 3.58 7.138 10.71 0.99 Table 2: Achieved bandwidth in Mbit s for the single VCs using the OSU scheme 0 1 2 3 4 5 ....

R. Jain, S. Kalyanaraman, R. Viswanthan, and R. Goyal, "A sample switch algorithm," Tech. Rep. 95-178R1, ATM Forum, Feb. 1995.

.... accepted by the ATM forum primarily because of the higher hardware complexity and costs involved in the latter scheme [31] see [32] for a general survey of the various proposals) Several heuristic algorithms for computation of explicit ABR rates have since been proposed at the ATM forum [28] [17]. Most other approaches ( 3] 22] 19] 24] 25] 12] 35] use fluid models of the network. Even though of obvious practical interest, very little is available in terms of stochastic control approaches in the queueing framework largely because they lead to analytical intractability when ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum/95-0178, Feb 1995.

....the right is reserved to add to, modify, or withdraw statements contained herein. Dynamic Averaging Interval Algorithm for ERICA ABR Control Scheme This contribution proposes a dynamic AI (averaging interval) version of ERICA (Explicit Rate Indication for Congestion Avoidance) scheme [1, 2, 3, 4]. ERICA algorithm shows different performance at different AI values and different configurations, which is described in the next section with simulation results. A new dynamic AI algorithm and its numerical results are presented in section 2. 1 Simulation of ERICA scheme Source and destination ....

Raj Jain, Shiv Kalyanaraman, Ram Viswanathan, and Rohit Goyal. A Sample Switch Algorithm. ATM Forum/95-0178R1, February 1995.

....by sending flow control indications to the traffic sources which are contending for the resources. There have been a number of schemes proposed to deal with this problem. We study some of recent proposals which are enhanced versions of PRCA (proportional rate control algorithm) EPRCA, ERICA, CAPC [4, 3, 7, 5]. 2. Congestion Control Schemes for ABR traffic Briefly, the differences between the baseline PRCA algorithm [1] and these enhanced versions are that there are periodic forwarding RM (resource management) cells from source to destination systems and all user cells are sent with EFCI (explicit ....

Raj Jain, Shiv Kalyanaraman, Ram Viswanathan, and Rohit Goyal. A Sample Switch Algorithm. ATM Forum/95-0178R1, February 1995.

....transient overload on the network. Motivated by a desire to keep queues small, RIF is often chosen to be small (we make observations on this typical choice later) 2. 2 The Switch Allocation Algorithm There are several switch algorithms proposed for computing the rate to be allocated to a VC [3, 13, 9, 7]. Switches compute an allocated rate for each VC i, based on its requested rate (value in the ER field) A i . VCs are classified as being either satisfied (in set S) or bottlenecked . The capacity C of the link is allocated to bottlenecked VCs as: AB = C Gamma P i2S A i N Gamma jjSjj ....

....the contribution to the queue by each flow (that per VC queueing provides) and the task of managing the queue becomes more difficult. Another alternative is to allocate only a portion of the actual available link capacity to the requesting VCs, and maintaining a reserve for draining the queue [7]. Firstly, the capacity available is potentially changing frequently, and even maintaining a reserve does not guarantee feasibility at all times. Secondly, we believe a goal of utilizing all of the available bandwidth is a desirable one, especially when there is little or no queueing in the ....

[Article contains additional citation context not shown here]

Jain, R., Kalyanaraman, S., Vishwanathan, R., Goyal, R., "A Sample Switch Algorithm", AF-TM 95-0178R1, ATM Forum Traffic Management Working Group, Feb. 1995.

....or CI bit to indicate the congestion state are referred as binary switches. Switches that modify the ER field are referred as ER switches. In the following we briefly describe the EFCI binary switch and the ERICA ER switch that we have used in the simulations carried out in this paper. See [3] and [4] for a complete description of the switch mechanisms. In [9] a performance comparison of different switch mechanisms is done. EFCI is the simplest switch mechanism. It monitors the queue length of the buffer and if it is larger than a threshold Q th the switch is considered congested. During the ....

....In an ABR connection each time a source transmits an RM cell, it sets the CCR field to the last computed ACR. CCR may be used by the switches as an approximate value of the transmission rate of the source, and several switch mechanisms use it in the computation of the ER feedback (e.g. ERICA [4] and EPRCA [3] At any time the CCR conveyed by a forward RM cell received at the measurement point should not be higher than the expected rate computed by the conformance algorithm (referred as PACR in the algorithm of figure 2) Transmission at a CCR higher than the ACR would imply exceeding ....

Raj Jain et al., "A Sample Switch Algorithm", ATM Forum contribution Number 950178R1, February 1995.

....more detailed description of the required computations see [7] 2. 3 The OSU Congestion Avoidance Scheme This scheme was developed by Jain et al. 8] at the Ohio State University (OSU) In this scheme the switches measure their input rate over a fixed averaging interval that is counted in cells [9]. Thereby, the input rate is calculated as follows: Input rate = number of received cells length of measurement interval With this calculated input rate a load factor can be computed which can then be used to indicate the over or underload state of the switch. load factor = input rate Target ....

....This problem can be solved by using larger measurement intervals. However, this reduces the reactivity of the algorithm and leads to longer transient periods. Actually, in our tests we already used a measurement interval of 60 cells which is already twice as large as the suggested value in [9]. Throughput (Mbps) Time (sec) 2 3 4 5 6 7 8 9 10 11 12 13 0 0.2 0.4 0.6 0.8 1 VC 6 VC 5 VC 4 VC 3 VC 2 VC 1 VC 0 Figure 2: Average path throughput achieved with the OSU scheme Another problem with the OSU scheme is its unfairness towards short distance traffic during overload periods. We have ....

R. Jain, S. Kalyanaraman, R. Viswanthan, and R. Goyal, "A sample switch algorithm, " Tech. Rep. 95-178R1, ATM Forum, Feb. 1995.

....fed by greedy sources can be found. L. Cerd a, O. Casals A Simulation Study of Switching Mechanisms for ABR Service 3 DES SES Forward RM Cell Backward RM Cell Figure 1: Rate Based Control scheme Lately several switches with different degree of performance and complexity have been suggested [10], 6] 2] In our paper we have chosen three switching mechanisms, namely the EFCI, EPRCA and ERICA, that show the different degrees of performance and complexity that can be achieved. A differential point of our study with respect to previous ones is considering most of all the parameters and ....

....ER2) Store NER in VCs Table YES NO NO NO Cell Tx YES Input Rate, IR=N T Overload Factor, OF=IR TCR Fair Share, FS=TCR VCs seen Count=0, VCs seen=0, T=0 YES NO YES Figure 4: ERICA Switch mechanism 3. 3 ERICA Switch The Explicit Rate Indication for Congestion Avoidance (ERICA) algorithm [10] is a proposal that tries to keep the queue length low and achieve a max min fairness. The switch mechanism is described in the flow chart of figure 4. A main difference with the previous switch mechanisms is the detection of the congestion state. In the previous mechanisms this detection is based ....

Raj Jain et al., "A Sample Switch Algorithm", ATM Forum contribution Number 950178R1, February 1995

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum95-( , 1995.

....rate switches for local area network (LAN) topologies. Since LANs have short feedback loops, some properties of the ABR control mechanisms may not be clearly observed. This paper studies the effect of unidirectional TCP traffic over ABR with Explicit Rate (ER) switches running the ERICA algorithm [2]. We use wide area network (WAN) configurations with finite buffers at the switches. Local area networks (LANs) have short feedback loops and the ABR control is found to be effective in LAN cases. We study the effect of timer granularity, switch buffer capacity, tail drop at switches, variable ....

....for very low values of TBE. On the other hand, TBE exponentially reduces ACR and may rate limit the TCP source, leading to drop in throughput. 3.2 An ABR Explicit Rate Switch Scheme : ERICA In this section, we present a brief overview of the ERICA switch algorithm. More details can be found in [1, 2]. Explicit Rate Indication for Congestion Avoidance (ERICA) is a simple switch scheme that allocates bandwidth fairly with a fast response. The scheme consists of using a Target Utilization of, say, 90 . The Target Rate is then set at: Target Rate = Target Utilization Theta Link Rate Since VBR ....

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A Sample Switch Algorithm, " ATM Forum 95-0178R1, February 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A Sample Switch Algorithm," ATM Forum 95-0178R1, February 1995. Available through http://www.cis.ohio-state.edu/~jain/

....to adjust their rates by the load factor. This allows the scheme to keep the three distinguishing feature while making the overhead independent of number of VCs. Newer versions of the OSU scheme, named ERICA (Explicit Congestion Indication for Congestion Avoidance) and ERICA are count based [22, 23]. 8.4 Congestion Avoidance using Proportional Control (CAPC) Andy Barnhart from Hughes Systems has proposed a scheme called Congestion Avoidance using Proportional Control (CAPC) 2] In this scheme, as in OSU scheme, the switches set a target utilization slightly below 1. This helps keep the ....

R. Jain, S. Kalyanaraman, and R. Viswanathan, "A Sample Switch Algorithm," AF-TM 95-0178R1, February 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A sample switch algorithm," ATM Forum Contribution 95-0178, February 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum 95-0178R1, 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum95-0- , 1995.

No context found.

R. Jain et al., "A Sample Switch Algorithm," ATM Forum Contribution 95-0178R1, 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan and R. Goyal, "A Sample Switch Algorithm", ATM Forum 95-0178R1, February 6-10, 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum 95-0178R1, 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal. A sample switch algorithm. ATM Forum 95-0178R1, 1995.

No context found.

Raj Jain, A sample switch algorithm, ATM Forum 95-0178, Feb. 1995.

No context found.

R. Jain, S. Kalyanaraman, R. Viswanathan, and R. Goyal, "A Sample Switch Algorithm, " AF-TM 95-0178R1, February 1995.

No context found.

Raj Jain, Shiv Kalyanaraman, Ram Viswanathan, and Rohit Goyal. A Sample Switch Algorithm. ATM Forum/95-0178R1, February 1995.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC