Performance Aspects of Some Switching Networks

Zbigniew Hulicki

Department of Telecommunication, AGH University of Science and Technology, ul. Czarnowiejska 78, Krakow, Poland

Keywords: Switching Fabric, Multistage Interconnection Network (MIN), Scalable High-Performance Router.

Abstract: The intention of this paper is to examine the performance and reliability aspects of switching fabrics which

are used for scalable high-performance routers. Topology and capabilities of switching fabrics are

discussed, followed by an examination of the performance vs.– reliability trade-off in diverse scenarios of

possible failures. It has been shown that the switching fabric based on the PM2I type multistage

interconnection network (MIN) outperforms those based on the cube type MIN under any failure scenario.

The simulation results presented should be helpful in predicting the performance vs.– reliability trade-off

before actual fabrication of the switching fabric.

1 INTRODUCTION

Routers and switches are viewed as the most critical

parts of the current communication infrastructure

and will be also needed to provide fast and efficient

communication in next-generation networks

Recently there have been various efforts to design an

efficient optical switch for use in a high-speed router

(Hossfeld et al., 2009). Besides, there is an immense

interest in designing a simple and high performance

switch which would satisfy the demands of an

entirely new scenario for emerging broadband

services.

In the past few decades, a number of switch

fabrics have been proposed in literature and used in

practical implementations to interconnect key

components in routers, such as routing engines and

line cards (Lien et al., 2010; Rongsen and Delgado-

Frias, 2007). Many of the proposed switch designs

have been based on multistage interconnection

networks (MINs). Various solutions, i.e. single or

multiple panel (replicated), can be used in design of

MIN based switches (Aulakh, 2006).

The purpose of this paper is to examine the

issues dealing with the performance and reliability

of single plane MINs used as switch fabrics for

scalable high-performance routers and/or OXCs.

Topology and capabilities of switching fabric will be

discussed first, followed by a specification of the

switch model and its performance measures. Then,

the performance and dependability of MIN based

switch fabrics will be examined, taking into account

diverse scenarios of possible faults. Lastly, there are

some conclusions and remarks regarding the trade-

offs between switch characteristics

2 TOPOLOGY OF SWITCHING

FABRIC

In selecting the architecture of switching fabric, four

design decisions can be identified: operation mode,

control strategy, switching method, and topology of

MIN. The topology of a switching network is a key

factor in determining a suitable fabric architecture

(Skalis and Mhamd, 2010). MINs proposed for the

switch fabrics are usually constructed using small

crossbar switches organized in stages and may have

a uniform or non-uniform connection pattern

between stages.

SEs can be buffered or non-

buffered, but the use of self-routing MINs are

favored because small delays can be achieved.

Generally, the topologies of MINs tend to be regular

and can be grouped into two categories: static or

dynamic.

The static MINs are simple to build and

expand, but fail even in the presence of a single

fault. On the other hand, a dynamic MIN is able to

reroute the data through a fault free path if the

regular path is faulty or busy. Because the reliable

operation of a MIN is an important factor in overall

router performance, it is important to design a

switching network that combines full connection

Hulicki Z..

Performance Aspects of Some Switching Networks.

DOI: 10.5220/0004428200900093

In Proceedings of the 3rd International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH-2013),

pages 90-93

ISBN: 978-989-8565-69-3

 2013 SCITEPRESS (Science and Technology Publications, Lda.)

capability – in spite of faults – with a slightly lower

performance.

A number of diverse fabric solutions which offer

redundant paths between the input and output ports

have been proposed to provide fast, efficient and

reliable communications at a reasonable cost.

However, no single solution network is generally

preferred because the cost-effectiveness of a

particular design varies with such factors as its

application, the required speed of data transfer, the

actual hardware implementation of the switching

network, the number of input/output ports, and the

construction cost.

As far as MINs are concerned, they can be built

using a variety of structures. Two significant

examples of topological structures based on the

single plane design are switching networks based on

the cube

and PM2I interconnection functions (Lien

et al., 2010). Therefore, there is an open question

about the performance vs.-reliability trade-off that a

multiple-path structure of the PM2I network

might

offer, for various distributed control algorithms and

decision rules, when compared to a cube-type

network.

3 THE SWITCH MODEL

To evaluate the properties of a fabric based on a

given switching network (MIN), the following

assumptions are made about the operation and the

environment of the interconnection network: the

single plane design of switching fabric is taken into

account; the slotted traffic source model is used with

a uniform random distribution of packet (cell)

destinations (because, in this contribution, we are

focused on the performance vs. reliability trade-off);

the switching network is operated synchronously,

meaning that the packets are transmitted only at the

beginning of a time slot given by the packet clock

and each input link is offered the same traffic load;

the buffering is external to the switch fabric, i.e.

non-buffered switch fabrics formed from non-

buffered SEs are under consideration. Therefore, the

queuing effects are unaccounted in this model. All

output ports perform the function of a perfect sync

and a conflict is said to occur if more than one

packet arrives at the same output link at the same

time.

Two classes of packet switched MINs, one with

redundant interconnection paths and one without are

under consideration (i.e. the cube-type class of MINs

known as unique-path networks and the PM2I class

of MINs which have multiple paths between a given

network input-output pair). It is clear that faults in

the switching network often result in degraded

performance of a fabric. Hence, in order to capture

effects of physical failures on the operation of

switching fabric, generally three models are used:

the stuck at fault model, the link fault model, and the

SE fault model (Aulakh, 2006). The link and SE

fault models will be used in this contribution. In the

former model, physical failures result in a faulty link

whereas, in the latter model in a faulty SE.

Moreover, in both cases the single and multiple fault

model versions will be considered. Because failures

occur at random, and because the combinations of

the type, number and location of faults can widely

vary, the influence of failures on the performance of

the switching fabric will be examined by simulation.

4 PERFORMANCE EVALUATION

The performance of switching fabric, based on both

of the aforementioned classes of packet switched

MINs, has been studied using simulated

experiments. A discrete time event-driven simulator

has been designed to carry out simulations. The

simulator uses the switch model and performance

measures described in the previous section. It

enables performance evaluation of the fabric under

different traffic and/or interconnection patterns,

including both uniform and non-uniform traffic

patterns, as well as an evaluation of the fabric’s

dependability under diverse fault models, including

combinations of the type, number and location of

faults. Moreover, in order to stop packets from

entering the switching fabric once its resources are

exhausted, the switching network of a router often

employs a backpressure mechanism (Lien et al.,

2010). Therefore, except for non-buffered MINs, the

simulator also allows for a modification to the

switching fabric, i.e. an implementation of buffering

(introduced at the input/output of the switching

network or to the SEs) or using different types of

SEs.

This section presents only selected results from

the simulation experiments because the results point

to a similar trend and the packet loss (or packet drop

rate) can serve as a good indicator of the

performance vs. reliability trade-off. The

simulation results have been shown and compared in

a few figures (Figs. 1, 2, 3, 4, and 5). Each data

value given in these figures is the result after

100,000 clock cycles in the simulation, where this

number of cycles is found to yield steady-state

outcomes (95% conf. lev.).

PerformanceAspectsofSomeSwitchingNetworks

Figure 1: Comparison of packet loss in a switch with

faulty SE.

Figure 2: Comparison of packet loss in a switch with

faulty SE.

The packet loss versus offered load under a

single SE fault for both the cube type (GC) and the

PM2I type () MIN is demonstrated in Fig. 1 and

Fig. 2 respectively. One can observe a qualitatively

similar nature of the performance of both switching

networks. Besides, it is clear (cf. Fig. 1 and 2) that

faultiness of a SE has crucial impact on the switch

performance. In both cases, the faulty SE determines

the degradation to the switch performance.

However, in the PM2I type MINs this effect will

depend on the stage location of the faulty SE. The

packet loss is higher if the faulty SE is located in the

output or input stage of the switching network (),

and the loss is lesser if the faults of SEs occur in the

inner stages of the MIN. In the cube-type MINs, the

faulty SE exhibits substantial impact to the

performance regardless of the fault locations.

Since faults have a strong tendency to happen in

continuous (or nearby) areas (Rongsen and Delgado-

Frias, 2007), the next tests had to consider the

multiple fault model scenario. The packet loss

versus offered load under a double SE fault for the

modified cube type (ESC) MIN is shown in Fig. 3.

It is observed that the location of the faulty SEs

determines the degradation of switching fabric

performance. A double SE failure has a more

detrimental effect on the performance (packet loss)

if defects of SEs happen in the inner stages of the

MIN, but the worst case is when the faults of SEs

happen in nearby areas.

Figure 3: Comparison of packet loss in a switch with

multiple faults.

Figure 4: Packet loss vs. traffic load in the MIN based

switch with faulty link.

On the other hand, there are SE and link failures

inside the switching fabrics. Therefore, it was

interesting to know what the effects would be of link

failures on switching fabric performance. In

consecutive tests, the impact of single faults has

been evaluated by randomly designating one link to

be faulty. The packet loss versus offered load under

a single link failure in different locations is depicted

in Fig. 4, and the results under a fault-free situation

are also included for comparison. It is observed that

a single link failure also leads to the performance

degradation; however, its impact is not as

detrimental as it was for an SE failure. The location

of the faulty link has (practically) small effect on the

performance degradation as well. Moreover, the

effects of link failures in the switching fabric based

on the PM2I type () MIN are different. One can

observe (cf. Fig. 5) more detrimental effects on

switching fabric performance when a straight link is

faulty and a failure occurs between the input and

output stage of switching network. Irrespective of

the location, other link failures have a negligible

SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand

Applications

Figure 5: Packet loss vs. traffic load in a switch with faulty

link in different location.

effect on performance degradation. Such capabilities

of the () switching network result from its

architecture which already has built-in redundancy

(multiple paths between a given network input-

output pair in the PM2I type MINs) that provides

fault tolerance. Furthermore, it is also clear (cf. Fig.

3 and Fig. 4 and/or Fig. 2 and Fig. 5) that an SE

failure has more detrimental effects on switching

fabric performance than a link failure. One could

expect this because SE hardware is much more

complex and, therefore, more prone to faults than

the internal link connections.

4 CONCLUSIONS

This paper presents the analysis of the performance

and reliability of single panel multistage switching

fabrics. Two classes of packet switched MINs, those

with redundant interconnection paths (the PM2I type

MINs) and those without (the cube type MINs), have

been examined by simulating performance and

reliability conditions. Simulations under different

fault model scenarios have shown that the

performance of multistage switching fabrics

deteriorates after an increase in load. However, the

switch based on the PM2I type MIN outperforms

that based on the cube type MIN under any failure

scenario. Moreover, the simulation revealed that

PM2I type multistage switching fabrics are highly

fault tolerant against internal link or SE failures.

This property can be preferable for scalable core

routers or backbone OXCs. Such single panel

multistage switching fabrics seem to be an attractive

alternative to the multiple panel architecture of

switching networks. Furthermore, presented results

of the simulation analysis should be also helpful in

predicting the performance vs. reliability trade-off

before actual fabrication of the switching fabric.

REFERENCES

Aulakh, N. S., 2006. ‘Reliability analysis of mux-demux

replicated multistage interconnection networks’,

Experimental Techniques, July/Aug. 2006, vol. 30,

(4), pp. 19–22.

Hossfeld, T., Leibnitz, K., Nakao, A., 2009. ‘Modeling of

modern router architectures supporting network

virtualization’, GLOBECOM Workshops, 2009 IEEE,

pp. 1–6.

Lien, Ch.-M., Chang, Ch.-S., Cheng, J., Lee D.-S., Liao,

J.-T., 2010. ‘Using banyan networks for load-balanced

switches with incremental update’, GLOBECOM

Workshops, 2010 IEEE, pp. 1–6.

Rongsen, H., Delgado-Frias, J.G., 2007. ‘Fault tolerant

interleaved switching fabrics for scalable high-

performance routers’, IEEE Trans. Paral. Distrib.

Syst., vol. 18, no. 12, Dec. 2007, pp. 1727-1738.

Skalis, N., Mhamd, L., 2010. ‘Performance guarantees in

partially buffered crossbar switches’, GLOBECOM

Workshops, 2010 IEEE, pp. 1–6.

PerformanceAspectsofSomeSwitchingNetworks