Parallel simulation of TCP/IP using TeD

Some applications of simulation require that the model state be advanced in simulation time than the wall-clock time advances as the simulation executes. This faster than requirement is crucial, for instance, when a simulation is used as part of a real-time control system, working through the consequences of contemplated control actions, in order to identify feasible (or even optimal) decisions. This paper considers the issue of than real-time simulation of very large communication networks, and how this is accomplished using our implementation (in C++) of the scalable simulation framework (SSF). Our tool (called iSSF) uses hierarchical levels of abstraction, and parallelism, to achieve speedups of nearly four orders of magnitude, enabling real-time execution rates on large network models. We quantify the effects that choice of hierarchical abstraction has on the simulation time advance rate, and show analytically and empirically how changing the abstraction mix affects performance.

Parallel and distributed simulation tools are emerging that offer the ability to perform detailed, packet-level simulations of large-scale computer networks on an unprecedented scale. The state-of-the-art in large-scale network simulation is characterized quantitatively. For this purpose, a metric based on the number of packet transmissions that can be processed by a simulator per second of wallclock time (PTS) is used as a means to quantitatively assess packet-level network simulator performance. An approach to realizing scalable network simulations that leverages existing sequential simulation models and software is described. Results from a recent performance study are presented concerning large-scale network simulation on a variety of platforms ranging from workstations to cluster computers to supercomputers. These experiments include runs utilizing as many as 1536 processors yielding performance as high as 106 million PTS. The performance of packet-level simulations of web and ftp traffic, and denial of service attacks on networks containing millions of network nodes are briefly described, including a run demonstrating the ability to simulate a million web traffic flows in near real-time. New opportunities and research challenges to fully exploit this capability are discussed.

Efficient calibration of microscopic car-following models for large-scale stochastic network simulators

In the context of heterogeneous networks and diverse communication devices, real person-to-person communication can be achieved with a universal personal identification (UPI) that uniquely identifies an individual and is independent of the access networks and communication devices. Hierarchical mobility management techniques (MMTs) have been proposed to support UPI. Traditional replication methods for improving the performance of such MMTs are threshold-based. We present optimal replication algorithms that minimize the network messaging cost based on network structure, communication link cost, and user calling and mobility statistics. We develop our algorithms for both unicast and multicast replica update strategies. The performance of these replication algorithms is studied via large scale network simulations and compared to results obtained from the threshold-based method.

In the context of heterogeneous networks and diverse communication devices, real person-to-person communications can be achieved with a universal personal identification (UPI) that uniquely identifies an individual and is independent of the access networks and communication devices. Hierarchical mobility management techniques (MMT) have been proposed to support UPI. Previously proposed profile replication algorithms for improving the performance of such MMT are either threshold-based or off-line. In contrast to an off-line algorithm that assumes perfect knowledge of user calling and mobility statistics, we propose in this paper on-line replication algorithms that decide whether to distribute new replicas or delete existing replicas after serving each request, all based on the input sequence seen so far. The performance of our on-line profile replication algorithms is studied via large scale network simulations.

Substantial evidence indicates that the time structure of neuronal spike trains is relevant in neuronal signal processing. Bio-inspired spiking neural networks are taking these results into account. Applications of these networks to low vision problems, e.g. segmentation, requires that the simulation of large-scale networks must be performed in a reasonable time. On this basis, we investigated the achievable performance of existing hardware platforms for the simulation of spiking neural networks with sizes from 8k neurons up to 512k neurons/50M synapses. We present results for workstations (Sparc-Ultra), digital signal processors (TMS-C8x), neurocomputers (CNAPS, SYNAPSE), small- and large-scale parallel-computers (4xPentium, CM-2, SP2) and discuss the specific implementation issues. According to our investigation, only supercomputers like CM-2 can match the performance requirements for the simulation of very large-scale spiking neural networks. Therefore, there is still. the need for low-cost hardware accelerators.

Traditional tools and software for social network analysis are seldom scalable and/or fast. This paper provides an overview of an R package called fastnet, a tool for scaling and speeding up the simulation and analysis of large-scale social networks. fastnet uses multi-core processing and sub-graph sampling algorithms to achieve the desired scale-up and speed-up. Simple examples, usages, and comparisons of scale-up and speed-up as compared to other R packages, i.e., igraph and statnet, are presented.

This paper presents a parallel algorithm for the simulation of large complex interconnect networks with frequency-dependent parameters. The equivalent transmission line model, which is suitable for time-domain simulation, is introduced for the modeling of dispersive interconnects. By means of the equivalent multi-port model, the interconnects are separated from other circuit elements, and thus makes the parallel simulation feasible. Based on the proposed technique, a program named FdSPICE is developed for parallel simulation of large interconnect networks. The accuracy and efficiency of FdSPICE are demonstrated by numerical examples.

While small topology simulations are important for validation and educational purposes, large-scale network simulations are a fundamental part of active networking research. Therefore, it is important for a network simulator to provide scalable and efficient solutions to execute these types of scenarios. Using standard sequential simulation techniques, large-scale topologies with substantial network traffic will require lengthy simulation execution times and a considerable amount of computer memory. Often, these execution times are too long for a networking researcher to run multiple simulations and collect significant data. Parallel and distributed simulation is one method that allows researchers to efficiently simulate these large topologies by distributing a single simulation program over multiple interconnected processors. To enable this scalable simulation methodology in ns-3, we formally present our distributed simulator, introduced in ns-3.8. This simulator uses the Message Passing Interface (MPI) standard and a conservative lookahead mechanism. Using the distributed simulator, we conducted a performance study using a large-scale point-to-point campus network scenario with a variable number of nodes distributed across several interconnected processors within a computer cluster. We show near-optimal improvements in simulation execution time are possible using the distributed simulator in ns-3.

Simulation of spiking neural networks — architectures and implementations

The reproduction of the brain ’sactivity and its functionality is the main goal of modern neuroscience. To this aim, several models have been proposed to describe the activity of single neurons at different levels of detail. Then, single neurons are linked together to build a network, in order to reproduce complex behaviors. In the literature, different network-building rules and models have been described, targeting realistic distributions and connections of the neurons. In particular, the Granular layEr Simulator (GES) performs the granular layer network reconstruction considering biologically realistic rules to connect the neurons. Moreover, it simulates the network considering the Hodgkin–Huxley model. The work proposed in this paper adopts the network reconstruction model of GES and proposes a simulation module based on Leaky Integrate and Fire (LIF) model. This simulator targets the reproduction of the activity of large scale networks, exploiting the GPU technology to reduce the processing times. Experimental results show that a multi-GPU system reduces the simulation of a network with more than 1.8 million neurons from approximately 54 to 13 h.

The computational complexity involved in modelling district heating (DH) networks impedes the integration of network operations into comprehensive DH system studies. We developed a flexible, accurate, and fast dynamic thermal simulation model utilising discrete event simulation (DES). This model is versatile, suitable for any tree-shaped DH network with a central heating plant and can estimate node temperatures and calculate pipe heat losses. The speed of the model is improved via using variable time steps and by incorporating two advanced techniques: lazy evaluation and a customised priority queue. To further improve the computational speed, we developed a technique to eliminate redundant sampling points. This model was tested and demonstrated excellent consistency with actual measurements. Remarkably, reducing sampling points can speed up the simulation by a factor of three without compromising the temperature accuracy. A 72-day simulation of a network with 102 pipes was completed within 0.219 s. Our findings highlight the significant potential of the DES model for large-scale dynamic network simulations and offer a promising solution for DH network simulations and system optimisation.

We are concerned with the simulation and optimization of large-scale gas pipeline systems in an error-controlled environment. The gas flow dynamics is locally approximated by sufficiently accurate physical models taken from a hierarchy of decreasing complexity and varying over time. Feasible work regions of compressor stations consisting of several turbo compressors are included by semiconvex approximations of aggregated characteristic fields. A discrete adjoint approach within a first-discretize-then-optimize strategy is proposed and a sequential quadratic programming with an active set strategy is applied to solve the nonlinear constrained optimization problems resulting from a validation of nominations. The method proposed here accelerates the computation of near-term forecasts of sudden changes in the gas management and allows for an economic control of intra-day gas flow schedules in large networks. Case studies for real gas pipeline systems show the remarkable performance of the new method.

To make good decisions, we need to be suitably informed. 'Good' and 'Suitably' in this case depend on the informational needs of the decision and the mechanisms of getting the information to the decision maker in time. The trade-offs in qualities, quantities, timeliness, impacts on other activities, and so on are infamously wickedly complex, and usually buried in a clutter of special circumstances, personality characteristics, environments unsuitable for study, and so on. Decision-making systems can be explored using case studies and exercises, but these are limited by the expense and time of using real people. A virtual simulator for large scale networks of communities can provide systems to examine that are not otherwise possible, while bearing in mind that simulators only partially reflect real systems. This paper describes a design for such a simulator framework that can be implemented on an ordinary desktop computer. We intend to use it to exercise and explore various ‘knowledge distribution strategies’ in order to understand and suggest information communication mechanisms for investigation in the real world, without expecting it to be complete enough to be prescriptive. We focus on military collaborations as suitably 'eXtreme' environments to exercise these communication mechanisms. Topics for further investigation include isolation, turnover and resilience.

On the performance of a hybrid network traffic model