Efficient real-time generation of a large class of bursty traffic streams (original) (raw)
Related papers
Journal of Communications and Networks, 2005
This paper presents an advanced architecture for a traffic generator capable of producing ATM traffic streams according to fully general semi-Markovian stochastic models. The architecture employs a basic traffic generator platform and enhances it by adding facilities for "driving" the cell generation process through high-level specifications. Several kinds of optimization are employed for enhancing the software's speed to match the hardware's potential and for ensuring that traffic streams corresponding to models with a wide range of parameters can be generated efficiently and reliably. The proposed traffic generation procedure is highly modular. Thus, although this paper deals with ATM traffic, the main elements of the architecture can be used equally well for generating traffic loads on other networking technologies, IP-based networks being a notable example.
A flexible and cost-effective ATM Traffic Generator
Performance Analysis of ATM Networks, 2000
In this paper a flexible burst-level ATM traffic generator is presented. It is a PC-based system with a careful allocation of functions between hardware (PC ISA board) and software, in a way that allows it to work on-line at fullspeed (155.52Mbps), on the one hand, and on the other, to be flexible enough to emulate a wide range of ATM traffic profiles. Up to 4 boards can be hosted by a single PC, each being able to generate up to 16 independent streams. Each stream consists of a continuous sequence of "traffic events" (burst-silence cycles), with each event being described by three parameters: the burst size, the silence duration and the inter-cell distance within the burst. The triplets, describing respective traffic events, are generated by software and downloaded on-line (through DMA) to the hardware; thus, in principle, any traffic model can be implemented, the limits being imposed only by the speed of the PC in relation to the total number of independent streams (up to 64) generated concurrently. For example, arbitrary distributions of the three traffic parameters (including experimental histograms) can be independently sampled; or correlated samples according to any law desired can be easily produced. Even further, real traffic can be emulated, provided that a monitoring device (e.g. a LAN traffic monitor) is connected to the PC feeding it with samples of the real traffic stream.
Enhanced Software Based Traffic Generator
This Paper provides an overview of the limitations imposed by software based traffic generator used to generate packet on the type of hardware as well as on the latency the operating system create for transmitting a packet. The paper also explains various stages through which a packet flows from the user-space until it is transmitted out from the hardware. Also discusses the architecture, design, implementation and performance between the socket type PF_PACKET, which is used by kernel based traffic generator PKTGEN to generate packets/traffic, also by most of software based traffic generators and a newly created socket type PF_ZERO, whose purpose is for traffic generation better than PF_PACKET by providing a way to generate higher traffic rates when compared to the existing mechanisms.
Experimental Results of Traffic Models for Burst Data and Voice Sources in Atm Networks
This paper presents an overview of the existing traffic models for burst data and voice sources in ATM, as well as some experimental results and comments about their implementations. The first paragraph is devoted to burst traffic generated by ON/OFF sources of constant throughput. The Matlab-based scheduler is able to determine the number of ON cells to the number of OFF cells ratio, for every burst, until the transmission process is completed. In the cases of multiple constant bit rate sources or a single source of variable bit rate, the superposition of several independent models previously discussed can be applied. This is method is not useful for voice traffic, possible solutions being offered by Markov-Modulated Poisson process model or by fluid source model. Experiments carried out gave the Matlab-based schedule of ATM cells departure, generated by S multiplexed sources.
A Distributed Multiplatform Architecture For Traffic Generation
This paper deals with the specification and realization of a synthetic traffic generator for IP traffic, called D-ITG (Distributed Internet Traffic Generator). We point our attention on the innovative architectural choices and the consequent interesting results. Indeed, thanks to the novel proposed distributed architecture, D-ITG reaches the highest performance over all the considered platforms and it shows interesting and innovative features in terms of supported protocols, available traffic patterns and meters type. In a heterogeneous network scenario, we think that D-ITG is an essential tool for network testing and planning activities. We compared our tool with several of the currently available and most widely adopted traffic generators and we found that D-ITG offers several improvements in terms of both functionalities and performance. The presented architecture and the comparative analysis shown in this work confirm our assumption.
Using LiTGen, a realistic IP traffic model, to evaluate the impact of burstiness on performance
2008
For practical reasons, network simulators have to be designed on traffic models as realistic as possible. This paper presents the evaluation of LiTGen, a realistic IP traffic model, for the generation of IP traffic with accurate time scale properties and performance. We confront LiTGen against real data traces 1 using two methods of evaluation. These methods respectively allow to observe the causes and consequences of the traffic burstiness. Using a wavelet spectrum analysis, we first highlight the intrinsic characteristics of the traffic and show LiTGen's ability to reproduce accurately the captured traffic correlation structures over a wide range of timescales. Then, a performance analysis based on simulations quantifies the impact of these characteristics on a simple queuing system, and demonstrates LiTGen's ability to generate synthetic traffic leading to realistic performance. Finally, we conduct an investigation for a possible model reduction using memoryless assumptions.
High Performance Internet Traffic Generators
The Journal of …, 2006
In the networking field, traffic generator platforms are of a paramount importance. This paper deals with the description of a distributed software platform for synthetic traffic generation over IPv4/v6 networks, called D-ITG (Distributed Internet Traffic Generator). We point our attention on the original architectural choices and evaluate the performance achieved by the platform. D-ITG supports several protocols and many traffic patterns. We tested our generation platform over different scenarios and compared it to many of the currently available, and most widely adopted, traffic generators. We found that D-ITG offers enhanced functionalities and improved performance.
Realistic and responsive network traffic generation
ACM SIGCOMM Computer Communication Review, 2006
This paper presents Swing, a closed-loop, network-responsive traffic generator that accurately captures the packet interactions of a range of applications using a simple structural model. Starting from observed traffic at a single point in the network, Swing automatically extracts distributions for user, application, and network behavior. It then generates live traffic corresponding to the underlying models in a network emulation environment running commodity network protocol stacks. We find that the generated traces are statistically similar to the original traces. Further, to the best of our knowledge, we are the first to reproduce burstiness in traffic across a range of timescales using a model applicable to a variety of network settings. An initial sensitivity analysis reveals the importance of capturing and recreating user, application, and network characteristics to accurately reproduce such burstiness. Finally, we explore Swing's ability to vary user characteristics, appl...
Development of a Modified Method of Network Traffic Forming
Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska
Based on the analysis of the statistical characteristics of heterogeneous network traffic of "Quadruple Play" mobile subscribers, it is shown that it cannot be represented by Poisson or Erlang distributions. It is shown that for such similar traffic, the rate of growth of the required buffer volume increases as the Hurst parameter increases. A method of adaptive traffic formation using the control of the length of intervals of the intensity of the arrival of data packets has been developed. Congestion management is carried out by changing the frequency of the marker generator (GM) on the basis of the results of forecasting the required bandwidth of the system and the required buffer size. The shaper adapts to changes in the length and instantaneous intensity of packet input in real-time.