Fluid-Flow and Packet-Level Models of Data Networks Unified Under a Modular/Hierarchical Framework: Speedups and Simplicity, Combined (original) (raw)
Related papers
Internet Performance and Control of Network Systems II, 2001
Fluid flow analytical models have been shown to be able to capture the dynamics of TCP flows and can scale well to solving for networks with a large number of flows. However, accurate closed form solutions are not yet available for wireless networks. Traditional packet-level discrete event simulations provide accurate predictions of network behavior, but their solution time can increase significantly with the number of flows being simulated. Integration of fluid flow models with packet-level simulators appears to offer significant benefits. In this paper, we describe an approach to integrate fluid flow models into QualNet, a scalable packet-level simulator. We validate the mixed model with detailed packet-level simulations for the scenarios considered in this paper. The execution time of the mixed model is significantly impacted by the frequency with which the analytical model must be solved in response to changes in the data rate at the interface of the packet-level and analytical models. We present a time averaging approach to mitigate this impact and present the results of the resulting tradeoff between prediction accuracy and model execution time.
Integrated fluid and packet network simulations
Proceedings. 10th IEEE International Symposium on Modeling, Analysis and Simulation of Computer and Telecommunications Systems, 2002
A number of methods exist that can be used to create sim- ulation models for measuring the performance of com- puter networks. The most commonly used method is packet level simulation, which models the detailed be- havior of every packet in the network, and results in a highly accurate picture of overall network behavior. A less frequently used, but sometimes
On the efficiency of fluid simulation of networks
Computer Networks, 2006
Performance evaluation of computer networks through traditional packet-level simulation is becoming increasingly difficult as networks grow in size along different dimensions. Due to its higher level of abstraction, fluid simulation is a promising approach for evaluating large-scale network models. In this paper we focus on evaluating and comparing the computational effort required for fluid-and packet-level simulation. To measure the computational effort required by a simulation approach, we introduce the concept of ''simulation event rate'', a measure that is both analytically tractable and adequate. We identify the fundamental factors that contribute to the simulation event rate in fluid-and packet-level simulations and provide an analytical characterization of the simulation event rate for specific network models. Among such factors, we identify the ''ripple effect'' as a significant contributor to the computational effort required by fluid simulation. We also show that the parameter space of a given network model can be divided into different regions where one simulation technique is more efficient than the other. In particular, we consider a realistic large-scale network and demonstrate how the computational effort depends on simulation parameters. Finally, we show that flow aggregation can effectively reduce the impact of the ripple effect and that the ripple effect has less impact when simulating the WFQ scheduling policy.
A Study of Networks Simulation Efficiency: Fluid Simulation vs. Packet-level Simulation
2001
Network performance evaluation through traditional packetlevel simulation is becoming increasingly difficult as today's networks grow in scale along many dimensions. As a consequence, fluid simulation has been proposed to cope with the size and complexity of such systems. This study focuses on analyzing and comparing the relative efficiencies of fluid simulation and packet-level simulation for several network scenarios. We use the "simulation event" rate to measure the computational effort of the simulators and show that this measure is both adequate and accurate. For some scenarios, we derive analytical results for the simulation event rate and identify the major factors that contribute to the simulation event rate. Among these factors, the "ripple effect" is very important since it can significantly increase the fluid simulation event rate. For a tandem queueing system, we identify the boundary condition to establish regions where one simulation paradigm is more efficient than the other. Flow aggregation is considered as a technique to reduce the impact of the "ripple effect" in fluid simulation. We also show that WFQ scheduling discipline can limit the "ripple effect", making fluid simulation particularly well suited for WFQ models. Our results show that tradeoffs between parameters of a network model determines the most efficient simulation approach.
Hybrid packet/fluid flow network simulation
Seventeenth Workshop on Parallel and Distributed Simulation, 2003. (PADS 2003). Proceedings., 2003
Packet-level discrete-event network simulators use an event to model the movement of each packet in the network. This results in accurate models, but requires that many events are executed to simulate large, high bandwidth networks.
FluidSim: a tool to simulate fluid models of high-speed networks
Performance Evaluation, 2001
In this paper, we present a tool for the simulation of fluid models of high-speed telecommunication networks. The aim of such a simulator is to evaluate measures which cannot be obtained through standard tools in reasonable time or through analytical approaches. We follow an event-driven approach in which events are associated with rate changes in fluid flows. We show that under some loose restrictions on the sources, this suffices to efficiently simulate the evolution in time of fairly complex models. Some examples illustrate the utilization of this approach and the gain that can be observed over standard simulation tools.
A loss-event driven scalable fluid simulation method for high-speed networks
Computer Networks, 2010
Increase of size and bandwidth of computer network posed a research challenge to evaluate proposed TCP/IP protocol and corresponding queuing policies in this scenario. Simulation provides an easier and cheaper method to evaluate TCP proposals and queuing disciplines as compared to experiment with real hardware. In this paper, problem associated with scalability of current simulation method for high-speed network case is discussed. Hence, we present a scalable time-adaptive numerical simulation driven by loss events to represent dynamics of high-speed networks using fluid-based models. The new method uses a loss event to dynamically adjust the size of a time step for a numerical solver which solves a system of differential equations representing dynamics of protocols and nodes' behaviors. A numerical analysis of the proposed protocol is discussed. A simple simulation of high-speed TCP variants is presented using our method. The simulation results and analysis show that the time-adaptive method reduces computational time while achieving the same accuracy compared to that of a fixed step-size method.
On the Validity of Fluid-dynamic Models for Data Networks
Journal of Networks, 2012
Simulators have been acknowledged as one of the most flexible tool in studying and evaluating network performance. Variuous approaches have been followed in data networks modeling. The present paper explores the range of applicability of a fluid-dynamic model, proposed in [5], able to describe the load evolution of a data network. Looking at intermediate time scale, it is assumed that packets are conserved, and hence the packets density obey to a conservation law. The assumption underlying the model is the behaviour of the information loss probability from which the flux is derived. Here the validity of the above hypothesis and the rules introduced to solve dynamics at nodes are discussed. In particular packet loss estimations using queueing models have been compared with the assumed loss probability. Moreover a TCP throughput simulation is proposed to evaluate the feasibility of the "tent" flux.
2002
In this paper we present a general hybrid systems modeling framework to describe the flow of traffic in communication networks. To characterize network behavior, these models use averaging to continuously approximate discrete variables such as congestion window and queue size. Because averaging occurs over short time intervals, one still models discrete events such as the occurrence of a drop and the consequent reaction (e.g., congestion control). The proposed hybrid systems modeling framework fills the gap between packet-level and fluid-based models: by averaging discrete variables over a very short time scale (on the order of a round-trip time), our models are able to capture the dynamics of transient phenomena fairly accurately. This provides significant flexibility in modeling various congestion control mechanisms, different queuing policies, multicast transmission, etc. We validate our hybrid modeling methodology by comparing simulations of the hybrid models against packet-leve...
A hybrid systems modeling framework for fast and accurate simulation of data communication networks
ACM SIGMETRICS Performance Evaluation Review, 2003
In this paper we present a general hybrid systems modeling framework to describe the flow of traffic in communication networks. To characterize network behavior, these models use averaging to continuously approximate discrete variables such as congestion window and queue size. Because averaging occurs over short time intervals, one still models discrete events such as the occurrence of a drop and the consequent reaction (e.g., congestion control). The proposed hybrid systems modeling framework fills the gap between packet-level and fluid-based models: by averaging discrete variables over a very short time scale (on the order of a round-trip time), our models are able to capture the dynamics of transient phenomena fairly accurately. This provides significant flexibility in modeling various congestion control mechanisms, different queuing policies, multicast transmission, etc. We validate our hybrid modeling methodology by comparing simulations of the hybrid models against packet-level simulations. We find that the probability density functions produced by ns-2 and our hybrid model match very closely with an L 1 -distance of less than 1%. We also present complexity analysis of ns-2 and the hybrid model. These tests indicate that hybrid models are considerably faster.