Anthony James McGregor

Simulation with real world network stacks

Network simulation is used widely in network research to test new protocols, modifications to existing protocols and new ideas. The tool used in many cases is ns-2. The nature of the ns-2 protocols means that they are often based on theoretical models that might not behave in the same way as real networks. This paper presents the network simulation cradle which allows real world network stacks to be used in a wrapper that allows the stacks protocols to be used in the ns-2 network simulator. The network stacks from the open source operating systems Linux, FreeBSD and OpenBSD are included in the simulation cradle as well as a stack designed for embedded systems, 1wIP. Our results show that ns-2s TCP implementations do not match observed behaviour from real machines in some respects and using the network simulation cradle produces results closer to real world network stacks.

Distributed OpenGL rendering in network bandwidth constrained environments

Display walls made from multiple monitors are often used when very high resolution images are required. To utilise a display wall, rendering information must be sent to each computer that the monitors are connect to. The network is often the performance bottleneck for demanding applications, like high performance 3D animations. This paper introduces ClusterGL; a distribution library for OpenGL applications. ClusterGL reduces network traffic by using compression, frame differencing and multi-cast. Existing applications can use ClusterGL without recompilation. Benchmarks show that, for most applications, ClusterGL outperforms other systems that support unmodified OpenGL applications including Chromium and BroadcastGL. The difference is larger for more complex scene geometries and when there are more display machines. For example, when rendering OpenArena, ClusterGL outperforms Chromium by over 300% on the Symphony display wall at The University of Waikato, New Zealand. This display has 20 monitors supported by five computers connected by gigabit Ethernet, with a full resolution of over 35 megapixels. ClusterGL is freely available via Google Code.

Validation of simulated real world TCP stacks

The TCP models in ns-2 have been validated and are widely used in network research. They are however not aimed at producing results consistent with a TCP implementation, they are rather designed to be a general model for TCP congestion control. The Network Simulation Cradle makes real world TCP implementations available to ns-2: Linux, FreeBSD and OpenBSD can all be simulated as easily as using the original simplified models. These simulated TCP implementations can be validated by directly comparing packet traces from simulations to traces measured from a real network. We describe the Network Simulation Cradle, present packet trace comparison results showing the high degree of accuracy possible when simulating with real TCP implementations and briefly show how this is reflected in a simulation study of TCP throughput.

Measured comparative performance of TCP stacks

This extended abstract present findings on measured TCP performance of a range of network stacks. We have found that there are significant differences between the TCP implementations found in Linux, FreeBSD, OpenBSD and Windows XP.

A comparison of system monitoring methods, passive network monitoring and kernel instrumentation

This paper presents the comparison of two methods of system monitoring, passive network monitoring and kernel instrumentation. The comparison is made on the basis of passive network monitoring being used as a replacement for kernel instrumentation in some situations. Despite the fact that the passive network monitoring technique is shown to perform poorly as a direct replacement for kernel instrumentation, this paper indicates the areas where passive network monitoring could be used to the greatest advantage and presents methods by which the discrepancies between results of the two techniques could be minimised.

Using the IEEE 802.11 Frame Check Sequence as a pseudo random number for packet sampling in wireless networks

Low power devices such as common wireless router platforms are not capable of performing reliable full packet capture due to resource constraints. In order for such devices to be used to perform link-level measurement on IEEE 802.11 networks, a packet sampling technique is required in order to reliably capture a representative sample of frames. The traditional Berkeley Packet Filter mechanism found in UNIX-like operating systems does not directly support packet sampling as it provides no way of generating pseudo-random numbers and does not allow a filter program to keep state between invocations. This paper explores the use of the IEEE 802.11 Frame Check Sequence as a source of pseudo-random numbers for use when deciding whether to sample a packet. This theory is tested by analysing the distribution of Frame Check Sequences from a large, real world capture. Finally, a BPF program fragment is presented which can be used to efficiently select packets for sampling.

Effect of multiplexing HTTP connections over asymmetric high-bandwidth-delay product circuits

The growth of the Internet is placing strain on the world wide telecommunications infrastructure. In particular it is no longer possible to purchase capacity on terrestrial cables to some parts of the world. To meet the growing traffic needs of the Internet some Network Service Providers are deploying asymmetric satellite connections. These providers must chose an architecture for the international component of the network. Depending on the architecture chosen the traffic on the international circuit might be made up of a large number of independent connections or a smaller number of connections carrying aggregated traffic.The appropriate approach is not immediately apparent because there are opposing performance factors and because the effect of asymmetric delay on TCP performance is not well understood. In this paper we describe a discrete event simulation of the effect of carrying multiplexed HTTP connections over as asymmetric high bandwidth delay circuit. We show that a high degree of multiplexing mitigates against TCPs bandwidth delay product limits but that using a TCP connection per HTTP request causes a significant increase in delay.

Using simple per-hop capacity metrics to discover link layer network topology

At present, link layer topology discovery methodologies rely on protocols that are not universally available, such as SNMP. Such methodologies can only be applied to a subset of all possible networks. Our goal is to work towards a generic link layer topology discovery method that does not rely on SNMP. In this paper, we will present a new link layer topology discovery methodology based on variable packet size capacity estimation. We will also discuss the problems that arose from preliminary testing where different brands of network cards affected the capacity estimates used to detect serializations. As a result, topologically equivalent links fail to be classified as such by the capacity estimation tool. To circumvent this issue, non-VPS methods of capacity estimation that utilise back to back packet pairs have been investigated as a calibration technique.

Voice over IP: the impact of RSVP

Routing voice over the Internet has drawn considerable attention over the past few years. The Internet has generally provided poor support for delay sensitive voice traffic, and thus new protocols have been proposed to give QoS guarantees. This paper reports on the performance of the public Internet between New Zealand and the United States, and simulates the effects that a guaranteed QoS woudl have.