Article Versions
Export Article
Cite this article
  • Normal Style
  • MLA Style
  • APA Style
  • Chicago Style
Research Article
Open Access Peer-reviewed

Analytical Modeling of RRM-ATM Switch for Linear Increment and Exponential Decay of Source Data Rate

Mohsen Hosamo , S. P. Singh, Anand Mohan
Journal of Computer Networks. 2017, 4(1), 56-64. DOI: 10.12691/jcn-4-1-6
Published online: September 5, 2017

Abstract

This paper describes analytical modeling of relative rate marking (RRM) switch for available bit rate (ABR) ATM service considering linear increment and exponential decay of data rate(s) of the source(s) to achieve faster congestion control of the network through resource management of the RRM switch. Theoretical performance of the switch has been evaluated in respect of link utilization factor and cell loss probability. It is shown that the switch achieves faster control over source(s) along with improved link utilization factor. This is particularly attractive for congestion control in ABR-ATM networks. The developed switch model can find potential application in the design of RRM switch specific to a network environment.

1. Introduction

Architecture and performance parameters of asynchronous transfer mode (ATM) switches 1, 2, 3 greatly influence the quality of service (QoS) of ATM networks. This has been the major driving force for continued interest in the development of variety of ATM switches having their inherent merits for specific applications. Essentially relative rate marking (RRM) switches achieve speed superiority over explicit forward congestion indication (EFCI) by making use of congestion indication (CI) and no increase (NI) bits of resource management (RM) cells for congestion control 4. Further, RRM switches are obvious choice and thus being extensively used in ATM networks as they can operate with finer resolution and provide a good trade off between the congestion control time and hardware complexity 6, 9. In spite of attractive features, RRM switches are not yet fully explored nor optimized under various operational network parameters. Mars et al. used an RRM switch to investigate the performance of transmission control protocol (TCP) connection 7, Lapsley and Rumsewicz 8 discussed advantages and disadvantages of using NI bit in the feedback loop to control the source data rate. Mischa Schwartz 10 discussed another feedback method for congestion control in broadband networks and Plotkin and Sydir proposed an implementation of RRM switch 11.

This paper bridges the gap of knowledge about performance of RRM switches by developing its mathematical model which provides theoretical foundation for RRM switch analysis. Theoretical analysis of maximum and minimum queue length for determination of Available Bit Rate (ABR) and resource management delays is presented followed by computation of queue length, time delay parameters, link utilization () and cell loss probability () for a given QoS.

2. Switch Modeling

Figure 1 shows the basic RRM switch architecture, which contains a multiplexer (MUX) and de-multiplexer (DMUX) at the input and output respectively and a shared memory used as queue () to store data cells of sources. The is an important shared resource of RRM switch whose status is monitored for source data rate control using feedback loop. Therefore the techniques used for implementation and its allocation have significant impact on the overall ATM performance which warrants development of an appropriate theoretical model for analysis of the switch and its effect on the ATM network performance.

2.1. Analogy with Fluid Flow

Considering the shared memory of Figure 1 as a bucket having an inlet-outlet in which the cells are poured as fluid that comes out from outlet. Assuming there are "N" sources each having same allowed cell rate (), the total rate of incoming cells " ” at time "t" can be given as “” where is delay time of forward resource management cells Considering that output (or service) rate is then fluid flow approximation of the can be represented as in Figure 2 which shows the total incoming cell rate , outgoing cell rate = (available link cell rate), net rate of increase/decrease in will be “[]". Therefore the rate of change of queue length “” at time can be written as and hence the value of at any instant can be expressed as

(1)
2.3. Analysis of Q(t) and ACR

The network model of the RRM switch having shared memory is shown in Figure 3 along with definition of the relevant parameters given in Table 1. Figure 4 gives the simplified data flow view of ATM network indicating forward and backward resource management cells and and actual data cells. Under operating conditions, any change in the rate of data cells affects the rate of and the bounds of the source rate would be, where and are the minimum and peak cell rates respectively. The steps in which the cell rate of sources can be increased is 12, where is a positive integer and this step size is called rate increase factor (). Therefore increase in cell rate at any instant would be and for it shall be , where is number of data cells between two . Therefore the elemental increase in allowed cell rate for one source can be given as:

(2)

Integrating equation (2) and using boundary condition at gives

(3)

Therefore the value of at any instant can be found by considering its bounds as

(4)

The variations of and computed using equations (1) and (4) respectively are plotted as function of time in Figure 5 from which it is seen that has six operational phases between its limiting values and as discussed below.

Phase (i): Initially at , the source starts sending data and starts building up after delay until the is filled up to and when =, is sent with CI=0 and NI=1 to fix the data rate of the source to after a time delay of. Assuming that is increasing linearly from to its upper limit (peak cell rate)=in the worst case then the value of can be written as

(5)

and for special case then

(6)

For , can be expressed as

(7)

Therefore using equation (7), can be expressed as

(8)

and for

(9)

Therefore the maximum time delay for change in cell rate from to can be obtained as

(10)

Phase (ii): When equals starts decreasing after lapse of time, which is needed by the (with CI=1, NI=0) to arrive at the source. Therefore can be written as

(11)
(12)

Using equation (5) and substituting the value of into equation (12), we have

(13)

For ,

(14)

Therefore the time required to change between and i.e. can be found using equation (15) as

(15)

The total time delay during which == constant and after which the source starts reducing can be expressed as

(16)

Phase (iii): When decreases from to, the continues to increase until it reaches. Therefore can be expressed as

(17)

or

(18)

where

(19)

Considering the curve of Figure 5, we have

Therefore

(20)

and

(21)

As the rate between two cells cannot be decreased by more than rate decrease factor () 12, referring Figure 4 the decrease in cell rate would be and corresponding rate decrease for Nrm data cells would be . Therefore rate of decrease in data cell rate for one source can be expressed as

(22)

Solving equation (22) we get

(23)

Thus the appropriate value of can be determined as

(24)

There are another three phases for the variation when it decreases from to

Phase (iv): When decreases exponentially from to, starts decreasing after time delay when (with CI=1, NI=0) arrives at the switch. Assuming, can be written as

(25)

For the value of is found from curve of Figure 5 as

(26)

and time needed to reduce the cell rate to is

(27)

Phase (v): When reduces to, starts increasing after delay when (with CI=0, NI=1) arrives at the source. Therefore the final expression for as shown in Figure 5 can be written as

(28)

Referring curve of Figure 5, we have

(29)

Therefore

(30)

The time during which remains constant at can be found as

(31)

Phase (vi): Finally when reaches the increases from to

Therefore

(32)

and since can be expressed as

(33)

thus

(34)

3. Analysis of Network Parameters

This section describes the analysis of “link utilization factor” and “cell loss probability” which are the important network parameters. These parameters have been determined using the proposed model. The link utilization factor can be computed from the knowledge of the status of at any instant. For the shared buffer is fully utilized but the buffer underflow occurs when, therefore the number of waste cells

(35)

Therefore the waste bandwidth i.e. number of cells wasted during one cycle of queue building and depletion over the time period can be expressed as

(36)

The cycle time of can be expressed as

(37)

and link utilization factor can be written as

(38)

From equation (37) it is clear that link utilization factor varies in the range “” to “”when the waste bandwidth “” changes due to change in network traffic condition from “” to. However, the cell loss occurs when that causes overflow of buffer resulting in loss of cells. Therefore the number of cells lost “” in one cycle time of can be expressed as

(39)

Therefore the cell loss probability “can be computed from

(40)
3.1. Analytical Results

The values of and have been computed using equations (10), (15), (19) and (32) considering up to 10 different sources (i.e. Computation of and has been carried out taking fixed values of bounds ( and), source parameters (,,,) and network parameters (,,, ). Figure 6 shows the analytical results of variation in and as function of considering example values of = 2000, =1000, ==150 cells/sec., =0.1 cells/sec., , ,=32, =1/16, =1/16,== 1 msec. These selected parameters are also used for determination of using equations (19) and (32) and is found as a function of ‘round trip time’ (RTT=, ranging from 0-100 msec) with the help of equations (38), (19) and (32). The variation in as function of is shown in Figure 7.

Variations of and are calculated using equations (38) and (40) respectively (utilizing the and expressions of equations (32) and (19)) as function of RTT as shown in Figure 8. and are calculated using equations (38) and (40) respectively as function of N in the range 1- 10 as shown in Figure 9. Variation of determined as function of using equations (40) and (19) is shown in Figure 10.

Referring Figure 6 it can be inferred that the proposed model of the RRM switch offers the benefit of reduced time delay between the instant when attains its maximum value (=) and the time when reaches to with increasing number of sources, however, exhibits inverse behaviour. This reduction in can be useful in faster control of source data rate for reducing cell loss probability due to buffer overflow. It may be seen from Figure 7 that increases rapidly for more than three sources. The link utilization () decreases moderately with increasing RTT and the cell loss probability () increases as in Figure 8. The variation of “” and “” for different values of “” is shown in Figure 9 from which it is clear that higher link utilization is achievable for more number of data sources, however, this is accompanied by affordable increase in which can of course be minimized by increasing as shown in Figure 10.

4. Conclusion

Modeling of RRM switch has been carried out to facilitate theoretical evaluation of the important parameters of the switch like length variation, link utilization factor, cell loss probability as function of round trip time for resource management / number of sources considering linear increment and exponential decay of source data rates. The expressions for different parameters have been developed and results of theoretical computations considering an illustrative example are presented. It is shown that the RRM switch implementation can achieve faster congestion control of the network along with improved link utilization factor which is particularly attractive for large number of sources. Further, the cell loss probability can be reduced by affording higher value of . The presented model can find potential application by using the developed equations for the design of RRM switch specific to a network environment.

References

[1]  Roland Schegg and Brigitte Stangl, Information and Communication Technologies in Tourism 2017, Proceedings of the International Conference in Rome, Italy, January 24-26, 2017.
In article      View Article
 
[2]  Mounir Frikha, Ad Hoc Networks: Routing, QoS and Optimization, Wiley-ISTE, 2010.
In article      View Article
 
[3]  Charles E. Spurgeon, Joann Zimmerman, Ethernet switches: An introduction to network design with switches, O'Reilly Media, 2013.
In article      View Article
 
[4]  Varma and Subir, Internet Congestion Control, Morgan Kaufmann, 2015.
In article      View Article
 
[5]  M Mars, A Bianco, R Cigno, and M Munafo, TCP over ABR in ATM networks with variable topology and background traffic, ATM'96 Workshop, Aug, pp. 25-27, 1996.
In article      View Article
 
[6]  Thomas Potsch, Future Mobile Transport Protocols: Adaptive Congestion Control for Unpredictable Cellular Networks, Springer Fachmedien Wiesbaden, 2016.
In article      View Article
 
[7]  M Mars, A Bianco, R Cigno, and M Munafo, TCP over ABR in ATM networks with variable topology and background traffic, ATM'96 Workshop, Aug, pp. 25-27, 1996.
In article      View Article
 
[8]  D Lapsley and M Rumsewicz, Improved buffer efficiency via no increase flag in EFCI flow control, ATM'96 Workshop, Aug, 1996.
In article      
 
[9]  Mischa Schwartz, Broad Band Integrated Networks, Department of Electrical Engineering, Colombia University, New York, N.Y. 1996.
In article      View Article
 
[10]  Varadharajan Sridhar and Debashis Saha, Recent Advances in Broadband Integrated Network Operations and Services Management, June, 2011.
In article      View Article
 
[11]  N T Plotkin and J J Sydir, The rate mismatch problem in heterogeneous ABR flow control, INFOCOM'97, Kobe, Japan, pp. 1306-1316, April 1997.
In article      View Article
 
[12]  A P Zwart, A fluid queue with a finite buffer and subexponential input, Adv. Appl. Probabil., vol. 32, pp. 221-243, Mar. 2000.
In article      View Article
 
[13]  ATM Forum/af-tm-0056.000, PROJECT: ATM Forum Technical Committee Traffic Management Working Group, April 1996.
In article      
 

Published with license by Science and Education Publishing, Copyright © 2017 Mohsen Hosamo, S. P. Singh and Anand Mohan

Creative CommonsThis work is licensed under a Creative Commons Attribution 4.0 International License. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/

Cite this article:

Normal Style
Mohsen Hosamo, S. P. Singh, Anand Mohan. Analytical Modeling of RRM-ATM Switch for Linear Increment and Exponential Decay of Source Data Rate. Journal of Computer Networks. Vol. 4, No. 1, 2017, pp 56-64. http://pubs.sciepub.com/jcn/4/1/6
MLA Style
Hosamo, Mohsen, S. P. Singh, and Anand Mohan. "Analytical Modeling of RRM-ATM Switch for Linear Increment and Exponential Decay of Source Data Rate." Journal of Computer Networks 4.1 (2017): 56-64.
APA Style
Hosamo, M. , Singh, S. P. , & Mohan, A. (2017). Analytical Modeling of RRM-ATM Switch for Linear Increment and Exponential Decay of Source Data Rate. Journal of Computer Networks, 4(1), 56-64.
Chicago Style
Hosamo, Mohsen, S. P. Singh, and Anand Mohan. "Analytical Modeling of RRM-ATM Switch for Linear Increment and Exponential Decay of Source Data Rate." Journal of Computer Networks 4, no. 1 (2017): 56-64.
Share
[1]  Roland Schegg and Brigitte Stangl, Information and Communication Technologies in Tourism 2017, Proceedings of the International Conference in Rome, Italy, January 24-26, 2017.
In article      View Article
 
[2]  Mounir Frikha, Ad Hoc Networks: Routing, QoS and Optimization, Wiley-ISTE, 2010.
In article      View Article
 
[3]  Charles E. Spurgeon, Joann Zimmerman, Ethernet switches: An introduction to network design with switches, O'Reilly Media, 2013.
In article      View Article
 
[4]  Varma and Subir, Internet Congestion Control, Morgan Kaufmann, 2015.
In article      View Article
 
[5]  M Mars, A Bianco, R Cigno, and M Munafo, TCP over ABR in ATM networks with variable topology and background traffic, ATM'96 Workshop, Aug, pp. 25-27, 1996.
In article      View Article
 
[6]  Thomas Potsch, Future Mobile Transport Protocols: Adaptive Congestion Control for Unpredictable Cellular Networks, Springer Fachmedien Wiesbaden, 2016.
In article      View Article
 
[7]  M Mars, A Bianco, R Cigno, and M Munafo, TCP over ABR in ATM networks with variable topology and background traffic, ATM'96 Workshop, Aug, pp. 25-27, 1996.
In article      View Article
 
[8]  D Lapsley and M Rumsewicz, Improved buffer efficiency via no increase flag in EFCI flow control, ATM'96 Workshop, Aug, 1996.
In article      
 
[9]  Mischa Schwartz, Broad Band Integrated Networks, Department of Electrical Engineering, Colombia University, New York, N.Y. 1996.
In article      View Article
 
[10]  Varadharajan Sridhar and Debashis Saha, Recent Advances in Broadband Integrated Network Operations and Services Management, June, 2011.
In article      View Article
 
[11]  N T Plotkin and J J Sydir, The rate mismatch problem in heterogeneous ABR flow control, INFOCOM'97, Kobe, Japan, pp. 1306-1316, April 1997.
In article      View Article
 
[12]  A P Zwart, A fluid queue with a finite buffer and subexponential input, Adv. Appl. Probabil., vol. 32, pp. 221-243, Mar. 2000.
In article      View Article
 
[13]  ATM Forum/af-tm-0056.000, PROJECT: ATM Forum Technical Committee Traffic Management Working Group, April 1996.
In article