Rastin Pries

Internet access traffic measurement and analysis

The fast changing application types and their behavior require consecutive measurements of access networks. In this paper, we present the results of a 14-day measurement in an access network connecting 600 users with the Internet. Our application classification reveals a trend back to HTTP traffic, underlines the immense usage of flash videos, and unveils a participant of a Botnet. In addition, flow and user statistics are presented, which resulting traffic models can be used for simulation and emulation of access networks.

On the Suitability of the Short Message Service for Emergency Warning Systems

The Global System for Mobile Communications (GSM) is the most popular standard for mobile phones in the world with about 1.57 billion customers. The short message service (SMS) in GSM allows amongst others the transmission of short text messages to mobile phones. In case of an emergency, the SMS can be used to warn a large number of individuals. However, the short message delivery is performed on best-effort basis and no quality of service (QoS) agreement is given, i.e., there is no guarantee how long it takes a message to reach the recipient or that the message will be successfully delivered. In this paper, we investigate the message loss probability and the transmission time by performing measurements with different scenarios in a public GSM network. Then, we identify problems according to the requirements for an emergency warning system. We present a solution based on the cell broadcast in GSM and show numerical results regarding the delay and the capacity of the SMS

On Traffic Characteristics of a Broadband Wireless Internet Access

Internet traffic measurements and traffic characterization are essential for managing and optimizing network infrastructures. The increasing number of wireless Internet users and the changing application demands require consecutive traffic measurements. Therefore, we have performed measurements of home users at a broadband wireless access service provider in order to reflect the current traffic characteristics. In this paper, we present the results of these measurements like application distributions as well as changing traffic characteristics caused by user demands and new services. The results could be used by a network service provider to optimize its network performance in order to give Quality of Service (QoS) guarantees for home users in its fixed wireless network.

An HTTP web traffic model based on the top one million visited web pages

The ever-changing behavior of HTTP web pages requires an adjustment of the web models used for simulation and benchmarking. In this paper, we present the statistical data of the one million most visited web sites. Using this data, we examine the changes in size and number of objects by comparing our findings with well-known web traffic models. The results show a trend towards large pages including multimedia content. In addition, todays web pages are created dynamically, i.e., that the content is downloaded from web servers spread all around the world. Finally, we discuss new web traffic models and present the parameters gathered from our measurements.

Traffic Measurement and Analysis of a Broadband Wireless Internet Access

The increasing broadband wireless Internet usage and the limited wireless resources require a careful network management and optimization of the wireless Internet Service Providers (ISPs). Unfortunately, those providers often just have statistics about the overall usage and limited knowledge about the detailed application distribution as well as the traffic char- acteristics. In this paper we present user and traffic characteristics measured at a broadband wireless Internet access. The results show that the applications change quickly but the general characteristics like packet size and TCP/UDP percentage have not changed during the last years. Index Terms—traffic measurements, traffic classification, home networks, BWA, WLAN

Traffic characterization of a residential wireless Internet access

Traffic characterization is an important means for Internet Service Providers (ISPs) to adapt and to optimize their networks to the requirements of the customers. Most network measurements are performed in the backbone of these ISPs, showing both, residential and business Internet traffic. However, the traffic characteristics of business and home users differ significantly. Therefore, we have performed measurements of home users at a broadband wireless access service provider in order to reflect only home user traffic characteristics.In this paper, we present the results of these measurements, showing daily traffic fluctuations, flow statistics as well as application distributions. The results show a difference to backbone traffic characteristics. Furthermore, we observed a shift from web and Peer-to-Peer (P2P) file sharing traffic to streaming applications.

Using buffered playtime for QoE-oriented resource management of YouTube video streaming

YouTube is the most important online platform for streaming video clips. The popularity and the continuously increasing number of users pose new challenges for Internet service providers. In particular, in access networks where the transmission resources are limited and the providers are interested in reducing their operational expenditure, it is worth to efficiently optimise the network for popular services such as YouTube. In this paper, we propose different resource management mechanisms to improve the quality of experience (QoE) of YouTube users. In particular, we investigate the benefit of cross-layer resource management actions at the client and in the access network for YouTube video streaming. The proposed algorithms are evaluated in a wireless mesh testbed. The results show how to improve the YouTube QoE for the users with the help of client-based or network-based control actions. Copyright

On the usability of OpenFlow in data center environments

Since its introduction OpenFlow has been used as an enabler for network experiments in a variety of fields. Although OpenFlow was initially only used in the research domain, the concept is now finding its way into data centers due to the relatively cheap hardware and high flexibility. In this paper, we take a look at the scalability and usability of OpenFlow in data centers. Based on data center traffic models and OpenFlow measurements, we evaluate whether OpenFlow is able to cope with the short flow inter-arrival times of the traffic models by means of simulation.

Comparative Study of the IEEE 802.16 Random Access Mechanisms

The WiMAX technology based on the IEEE802.16 standard is currently the most prospective candidate for broadband wireless access networks. One of the key issues is the design of the MAC layer, in particular the multiple access scheme. The IEEE 802.16 standard specifies different scheduling services with individual mechanisms for accessing the channel on the uplink. The non-real-time polling service and the best-effort service mainly rely on a contention mechanism to submit bandwidth requests to the base station. These two services are currently used for all types of traffic with unknown characteristics, i.e. typically all traffic except for some special VoIP connection with known codec. In this paper, we evaluate and compare the performance of the contention mechanisms for fixed and mobile WiMAX.

A queuing analysis of an energy-saving mechanism in data centers

The high energy costs for running a data center led to a rethinking towards an energy-efficient operation of a data center. Designed for supporting the expected peak traffic load, the goal of the data center provider such as Amazon or Google is now to dynamically adapt the number of offered resources according to the current traffic load. In this paper, we present a queuing theoretical model to evaluate the trade-off between waiting time and power consumption if only a subset of servers is active all the time and the remaining servers are enabled on demand. We develop a queuing model with thresholds to turn-on reserve servers when needed. Furthermore, the resulting system behavior under varying parameters and requirements for Pareto optimality are studied.