FAFC: Fast Adaptive Fuzzy AQM Controller For TCP/IP Networks

Transcription

1 : Fast Adaptive Fuzzy AQM Controller For TCP/IP Networks Yassine HADJADJ AOUL, Abdelhamid NAFAA, Daniel NEGRU, Ahmed MEHAOUA Laboratoire CNRS-PRISM 45, Avenue des Etats Unis Versailles, France {yana, anaf, dan, Abstract Recently, many Active Queue Management (AQM) algorithms have been proposed to address performance degradations of end-to-end congestion control. However, these AQM algorithms present weaknesses for stabilizing delays in heavy loaded networks. In this paper, we describe a novel adaptive fuzzy control algorithm to improve best effort TCP/IP Networks performance. Comparing to traditional AQM algorithms (, and others), our proposal avoids buffer overflows/underflows, and minimizes packets dropping. We propose an on-line adaptation mechanism that captures fluctuating network conditions, while classical AQM algorithms require static tuning. The algorithm stability is mathematically proven. Simulation results show that for the same link utilization, provides better performance than and. Keywords-component; AQM; QoS; Fuzzy; Adaptive; Multimedia. I. INTRODUCTION Today s Internet traffic is dominated by TCP flows. TCP uses several mechanisms to handle network congestion such as: AIMD (Additive Increase and Multiplicative Decrease), slow start, congestion avoidance, fast retransmit and fast recovery. Besides, streaming audio and video on the Internet is, presently, well accepted and largely deployed (e.g. CDN Content Delivery Networks ). There also eist more and more multimedia applications with integrated congestion control behavior known as TCP-Friendly (e.g. TFRC, DCCP). This rapid epansion underlines a new challenge for efficient handling of network traffic. Especially, there is a growing need to fairly regulate competing network flows while maintaining a bounded end-to-end latency. Real-time media communication is characterized by data that have strict delivery constraints (delays, jitter, packets loss, bandwidth and so on). For eample, when receiving a streaming media session, data that arrive late is useless (i.e. if a packet delay is too high and the playout deadline is missed, a late loss occurs). Therefore, it is necessary to design and to deploy network QoS mechanisms. There are three major approaches commonly used to ensure end-to-end QoS in networks: () Intserv based on firm perflow resource reservations, () Diffserv that provides statically perclass guarantees and (3) AQM (Active Queue Management) designed for router queue management. The first two techniques are basically established on flows distinction assumption, whereas AQM can fairly regulate network traffics without any flows discrimination. This is particularly prevalent in best effort networks where all streams have the same network access right. Several AQM techniques have been proposed for congestion avoidance [4]. Among these, we have: Random Early Detection [4] and its variants, and Proportional Integral Derivative controllers [8]. mechanisms are widely deployed on today s IP routers. These mechanisms are often disabled due to their drawbacks []. On the other hand, mechanisms emerge from control theory. They perform in a better way the control of end-to-end delays. Nevertheless, compared to, involves more queue length oscillation, which increases the delays fluctuation. Using a large queue size, in AQM techniques, leads usually to high packet delays. As pointed out in this paper, the main drawback of common AQM algorithms is that they still induce high packet loss ratio when the buffer size is not large. In addition, the classical AQM algorithms are sensitive to network parameters changes (e.g. number of flows), which have devastating consequences on router throughput and packets latency (late losses). Our main concern in this study is to design an AQM that maintains the queue size stability while minimizing dropped packets. There are two main contributions. First, we elaborate the design of an efficient Fuzzy-based queue controller through the use of heuristic information (from Fuzzy Control Theory [8]). Second, we introduce a fast on-line adaptive mechanism. It makes the AQM algorithm more robust towards network parameters variations, noises (i.e. non adaptive flows), modeling errors, and model nonlinearities. The rest of this paper is organized as follows: Section II relates previous works on the common AQM techniques issues. In Section III, we present our proposed fast adaptive Fuzzy controller (). Section IV is devoted to the performance evaluation and results analysis. Finally, we conclude our paper in Section V. II. RELATED WORKS The internet dynamics research show that the TCP congestion avoidance mechanisms, while essential and powerful, are not sufficient to avoid the congestion since it s detected only after a packet dropping. Moreover, this mechanism is inadequate in some circumstances since it often implies large packets delay []. Usually, we can prevent the high delay at network level by introducing two distinct mechanisms Queue Management and Scheduling Algorithms. The first technique, addressed in this paper, manages the length of packet queues by dropping the out of traffic (ecess load) or even by controlling the queue length [] [3]. The second technique orders the packets in the queues. These two mechanisms are related nevertheless they address different aspect

2 of QoS provisioning. In this section, we review AQM related works highlighting their QoS capabilities. A. Random Early Detection: The main role of the is to avoid network congestions by controlling the average queue length [4][5]. It permits also to maintain the system stability through adopting a Drop Tail behavior when the queue reaches a certain threshold. permits to determine the signature of the congestion (the transient congestion is accommodated by a temporary increase in the queue size and the baseline congestion is reflected by an increase in the computed average queue length). Besides this, the algorithm permits a statistical fairness among sources with different burstiness. recalculates the average queue length with a low pass filter named EWMA Eponential Weighted Moving Average. It is particularly useful to adjust the smoothness/aggressivity of. It was shown in [7] that is stable. However its performance is sensitive to the parameters tuning (specifically w q, min th and ma th ). From this fact, several researches were carried out for improving the basic algorithm. The most important variants is Adaptive- [6] that automatically tunes the parameters. Since the -based algorithms control the macroscopic behavior of the queue length (average), they often cause sluggish response and fluctuation in the antaneous queue length. As a result, an important variation in end to end delays is observed. As a consequence, and its variants provoke dramatic consequences on sensitive flows. B. Control-Theoretic to Flow Control The control theoretic techniques have been lately introduced in flow control and congestion avoidance [4]. These recently developed mechanisms outperform the eisting works by presenting formal proofs, whereas the classical approaches are rather informal behavior approaches; they give the means to eplicitly specify multi criterions performance. These alternative methods model the AQM algorithm as a feedback control system that tunes the router queue length as a plant variable. Q ref AQM Controller TCP Delay Queue Feedback Plant Fig.. Feedback control modeling of congestion control with AQM In the system above, Q ref represents the reference value of the queue length. The Plant represents a combination of subsystems such as TCP sources, TCP receivers, routers... and others. Among the contributions in the control-theoretic to flow control field, [8][] proposed the use of the controller ( is the widely used controller in Automatic area). In [5] a feedback compensation element was introduced in order to provide a more robust controller under time-varying network conditions. Beside that, in [6] a variable structure based control scheme was used in order to take into account the model uncertainties and the number of active TCP connections. These controllers permit better performance compared to s. For ance, the controller adds to the integral control (present in s through the EWMA Q computing) a proportional control for a faster response, and a derivative control for anticipated congestion avoidance. The theoretic control in the network management domain brings a new alternative permitting good performance in router throughput and better queue length management. Note that the plant model is strongly uncertain, nonlinear (see equation ), and subject to additive noises (the noise consist of non adaptive flows e. g. UDP). This leads to parameterization problems under realistic traffic environment characterized by intrinsic bursty nature and time-variable structure. From this fact, it is much suitable to deploy an adaptive/auto-tuned control that tackles the changing network conditions. III. : FAST ADAPTIVE FUZZY CONTROLLER FOR AQM Fuzzy Control provides a formal methodology for representing, manipulating and implementing a human heuristic knowledge for systems controlling. In fact, fuzzy control focuses on gaining an intuitive understanding of how to control the process. Contrary to the conventional control approaches, those focus on constructing a controller by using the system model. This permits to the fuzzy controller to be less sensitive to model uncertainties, nonlinearities and noises; it then allows controlling a more comple system. In this section, we first etract stability criterions through Lyapunov s stability analysis and proofs. This is an efficient way to guarantee a stable controller. Afterwards, we use the stability criterions to design a Mamdani-based model that better captures the network nonlinearities. Last, we propose a fast adaptation Fuzzy controller in order to preserve the performance in varying network conditions. A. Stability analysis Basically, the mathematical model is employed to find the most suitable control for a given system or to deduce the optimized parameters. In our work, we use the stability study so that we etract different criterions needed to design a correct shape of the fuzzy controller. Note that, the correct shape represents a control surface stabilizing the system. We use the Lyapunov s Direct Method for the stability analysis of the fuzzy controller [8]; it permits to obtain a stable plant using the deduced constraints. The Lyapunov s Direct Method is the most popular technique for the stability analysis of nonlinear systems. By replacing the marking probability P () t with the fuzzy control Φ (, ) in the model developed by Frank Kelly [5] and generalized to support multiple TCP sessions by Crowcroft and Oechslin [], we obtain the following system: & = M M ( C ) () & = ( ) Φ, τ τ M Where M is the number of TCP sessions, τ the RTTs, C the link capacity, the error queue length and the derivative of. We choose: T V ( ) V = = = ()

3 P So that V & ( ) V ( ) = & & = (3) By replacing () in (3) we obtain: ( C ) ( ) M M V & = ( ) Φ, (4) τ τ M Since we aim to prove asymptotic stability, it is sufficient to have V & ( ) <. It is then sufficient to find a fied β > such as: ( C ) M M ( ) < β τ τ M Φ, (5) While, we have two cases: > and <. If > we obtain: (, ) β M τ > (6) M ( C ) τ M Φ Else: (, ) β M τ < (7) M ( C ) τ M Φ n When B( h ) = { R : < h, h > } the system is asymptotically stable (this corresponds to a ball centered at the origin with a radius ofh ). Thus, the design of the fuzzy controller Φ( ), must respect the equations (6) and (7). Note that the theory doesn t say how large ish. B. Design of the Fuzzy Controller During the controller design, we constrained ourselves to the two inputs Proportional-Derivative Fuzzy Controller (MISO, Multi Inputs Single Output). The first input represents the tracking error e (compared to the desired queue length) and the second input de represents the derivative of the first input. We graphically design the fuzzy controller output P in accordance with the equations (6) and (7) in order to obtain a stable controller. The control surface of our PD-Fuzzy controller is as follows: de Fig.. The control surface of the fuzzy controller In our case, the fuzzy controller behaves similarly to the PD controller, with the eception that the linear PD controller is unable to achieve a nonlinear control surface (see figure ). In addition, e 5 5 we use different gains as in the PD controller [8]. In, we use a proportional gain superior to one (amplifying the input) in order to permit a very small tracking error (static error). We observe that a large proportional gain involves a fast convergence while making the system more oscillatory. The controller implements only these three obvious fuzzy laws: If e is high and de is high then P is high If e is medium and de is medium then P is medium If e is low and de is low then P is low The shape and range of the fuzzy membership functions should be tuned according to the mean number of active TCP connections. This simple choice works well for any number of TCP connections after adding our adaptive mechanism (eplained in the section below). C. Fast Adaptation Mechanism Fuzzy controllers give the way to control uncertain and noisy systems. However, the TCP model variation is complicated, making difficult the design of fuzzy controllers able to give high performance when the number of flows fluctuates. Hence, the use of an adaptive control is much suitable. One of the most popular approaches is to use equally space triangular membership functions and common rules. Then, find a technique to modify, tune or adapt them to the plant variation, the unmodeled dynamics, and the eternal disturbances. There are lots of adaptive Fuzzy controllers in literature [7]. Most of them require tremendous resources in term of memory requirements and computing time and are not adapted to real-time systems especially in the AQM case, where the router must handle thousands of packets per second. In our approach, we avoid the membership functions modification in order to use the memory saving scheme [8]. Our proposal just makes a global translation in the membership functions. This has a direct impact on the controller robustness when network conditions vary, by smoothly increasing/decreasing the probability dropping level. The translation mechanism is done thanks to the classical controller (see equation 8). We apply the EWMA function as input to tune the Fuzzy controller balancing. This will permit a smooth adaptation avoiding fluctuations in the controller response. K i P = K K s e p d (8) s K, p K d and K i represent respectively the proportional gain, the derivative gain and the integral gain. e represents the controller input while P represents the output (the balancing value). Q ref du dt EWMA G p G d Feedback Fig. 3. Feedback Adaptive control modeling of congestion control with AQM algorithm FC Plant Q

4 The gains G p and G d represent respectively the proportional gain and the derivative gain. The makes an additive action at the fuzzy controller output in order to perform the balancing. At the output of our controller a saturation block is added, which sets the dropping probability value between and. IV. PERFORMANCE EVALUATION A. Simulation Model In order to evaluate the performance and robustness of our proposal, we performed simulations using the topology depicted in figure 4. The bottleneck is located at the central router level that implements one of the three AQM algorithms:, or Adaptive Fuzzy (our proposal). Its capacity is Mbps (6 packets/s, default packet size is 5 bytes), and delay is ms. Each queue may contain up to 4 packets. However, the most performing queue length is of 8 packets. In our simulation, the number of TCP connections changes in accordance with the curve in figure 5. RTT =. s delay in the queue permits only good performance in respect to the throughput (see Table ). We can see in figure 7 that the three controllers permit good performance. However, gives a constant and maimal throughput value which is showed in figure 6 by the stability of the queue size. Note that the maimal throughput of our router is Mbps, which means that any traffic ecess involves a buffering at the router queue, and even higher drop rate when the buffer queue length is reached (figure 6). Instantaneous queue length (KB) S D 4 Router S Mbp D / & Adaptive Fuzzy Fig. 6. Instantaneous queue length ( q ) S N D M Fig. 4. Simulation Network Topology Number of TCP sessions Throughput (KB) Fig. 5. Variations in time of TCP sessions number ( M ) Note that the parameters of and are set as recommended respectively in [4] and in [8]. For the fuzzy controller, we choose k p = 3 and k d = in order to get more precised results. The static error is eliminated thanks to the adaptive control action. B. Eperimental Results Figure 6 presents the queue evolution, which shows that the controller regulates quickly the queue to the reference value despites the work-load variation (figure 5). Contrary to, the controller converges slowly and is very disturbed with the variation of the number of TCP sessions. The statistics in Table show that, in the case of the controller, the static error decreases and the queue length converges to the referenced value asymptotically with the stabilization of the work-load. Figure 6 shows that doesn t guarantee the stability of the queue; it Fig. 7. Throughput ( r ) From figure 7 and 8, we clearly see that the queuing delays in are constant. and controllers are sensitive to the number of session s variations. Besides, the fluctuation in queuing delay implies a deep deterioration of Multimedia communications that have strict requirements in term of end to end delay and jitter. So, in order to provide QoS guarantees, the elimination of the queuing delay fluctuation is a must. Our controller reaches this objective. Furthermore, from statistics tables, we see that induces a maimum delay of.4 s, while the and controllers reach. s each. Looking at figure, we see that the dropping probability in is not regular. It increases quickly when the number of TCP sessions increase (see figure 5). Moreover, we see clearly that the dropping probability value in is better than in.

5 Queuing delay (s) queue is proven thanks to Lyapunov theory. The obtained controller is very simple to implement and improves network performance in terms of delay, jitter and packet loss rate, which is very important for multimedia applications requirements. The simulation eperiments demonstrated that permits to stabilize quickly the queue length despites the variations of the TCP sessions. Moreover, the performance obtained demonstrated the s capacity to capture the network model nonlinearities. Future works will focus on etending to deal with heterogeneous traffics requirements, thus providing differentiated dropping probabilities in real networks. Dropping probability Fig. 8. Queuing delay ( d ) Fig.. Dropping probability ( p ) The algorithm shows much better performance than and controllers. In fact, the performances of and are sensitive to network configurations. In addition, and can make a buffer overflow in a transient period which results in packets losses. All these remarks are summarized in Table II, which briefly describes advantages and drawbacks of each AQM algorithm treated in this paper. TABLE I. CURVES STATISTICS min ma mean min Ma mean min Ma mean q r.e4.99e4.e4.e4.e4.95e4 d TABLE II. A CLASSIFICATION OF AQM ALGORITHMS IN TERMS OF OBJECTIVES Characteristics Router Throughput good good good Sensitivity to network parameters no yes yes Response very fast slow fast Queue stability stable not stable stable Delay small very large variable Buffer overflow/underflow no/no yes/yes yes/yes Steady state error <% - <% Control Type I V. CONCLUSIONS In this paper, we propose a new adaptive fuzzy controller for AQM,. It permits a very fast response compared to the classical adaptive controllers, and. The stability of the REFERENCES [] Braden, B., and al.: Recommendations on Queue Management and Congestion Avoidance in the Internet. IETF RFC 39 (April 998) [] Lin, D.: Internet Congestion Control: Cooperative End-System and Gateway Algorithms. Doctor of Philosophy in Computer Science thesis in Harvard University (May 998) [3] Floyd, S., and Fall, K.: Promoting the Use of End-to-End Congestion Control in the Internet. IEEE/ACM Transactions on Networking (May 999) [4] Floyd, S., and Fall, K.: Random Early Detection Gateways for Congestion Avoidance. IEEE/ACM Transactions on Networking (August 993) [5] Heying, Z., Baohong, L., and Wenhua, D.: Design of Robust Active Queue Management Algorithm Based on Feedback Compensation. ACM/SIGCOMM 3 (August 3) [6] La, R., J., Ranjan, P. and Abed, E., H.: Analysis of Adaptive Random Early Detection A. 8 th ITC (September 3) [7] Ohsaki, H., Murata, M., and Miyahara, H.: Steady State Analysis of the Gateway: Stability, Transient Behavior, and Parameter Setting. IEICE Transactions on Communications, vol. E85-B, pp. 7-5 (January ) [8] Yanfei, F., and Chuang, F. L.: Design a Controller for Active Queue Management. ISCC 3 (July 3) [9] Keshav, S.: A Control-Theoretic to Flow Control. ACM SIGCOMM 99 (September 99) [] Ryu, S., Rump, C., and Qiao, C.: A Predictive and Robust Active Queue Management for Internet Congestion Control. ISCC 3 (June 3) [] Keshav, S.: Congestion Control in Computer Networks. PhD Thesis, published as UC Berkeley TR-654 (September 99) [] Bitorika, A., and al. A Comparative Study of Active Queue Management Schemes. ICC 4 (June 4) [3] Sun, J., et al. PD-Controller: A New Active Queue Management Scheme. GLOBECOM 3 (December 3) [4] Ryu, S., Rump, C., and Qiao, C.: Advances In Internet Congestion Control. IEEE Communication Surveys Vol. 5 No. (3) [5] Kelly, F.: Mathematical Modeling of the Internet. Springer-verlag, Berlin () [6] Yan, P., Gao, Y., Özbay, H.: Variable Structure Control in Active Queue Management for TCP with ECN. (June 3) [7] Jamshidi, M.: Large-Scale Systems: Modeling, Control and Fuzzy Logic. Edited by Prentice Hall PTR, ISBN (997) [8] Passino, K., M., Yurkovich, S.: Fuzzy Control. Edited by Prentice Hall, ISBN X (998) [9] Le, L., Aikat, J., Jeffay, K. and Donelson Smith, F.: The effect of Active Queue Management on WEB Performance. SIGCOMM 3, ACM Press, ISBN: (3) [] Crowcroft, J. and Oechslin, P.: services using a weighted proportionally fair sharing TCP. ACM Computer Communications Review 8,