Characterisation of Internet Traffic Matrices. The benefits to the community of characterisation of Internet traffic matrices arise because this information is used in the design of efficient and reliable networks. This is a genuine example of using the data we have more cleverly, rather than throwing money at a problem. Better information will result in the Internet becoming more efficient, and therefore cheaper, and at the same time more reliable.
Ensuring the Robustness of IP Routing. The Internet is by its nature international, and so any such effort to model and understand it must be somewhat international. However, Australia has longer network paths to the rest of the world and the routing protocols' operation depends on this latency. This emphasizes why Australia must participate in such research: to ensure that Australian issues are considered when Internet protocols are designed. This project will directly deliver a more robust Int ....Ensuring the Robustness of IP Routing. The Internet is by its nature international, and so any such effort to model and understand it must be somewhat international. However, Australia has longer network paths to the rest of the world and the routing protocols' operation depends on this latency. This emphasizes why Australia must participate in such research: to ensure that Australian issues are considered when Internet protocols are designed. This project will directly deliver a more robust Internet for Australia, international exposure in an area of general interest, the potential for patents, and the development of commercially valuable expertise and measurement infrastructure.Read moreRead less
Delivering guaranteed quality of service over IP networks. The next revolution in the Internet will be to offer end-to-end Quality of Service (QoS) guarantees, in the form of maximum delay, delay variation and probability of packet loss, or minimum throughput. There are no existing techniques capable of delivering end-to-end QoS guarantees. This project will develop credit-based controls, instead of the usual rate-based controls, that can deliver these required QoS guarantees. These QoS guarante ....Delivering guaranteed quality of service over IP networks. The next revolution in the Internet will be to offer end-to-end Quality of Service (QoS) guarantees, in the form of maximum delay, delay variation and probability of packet loss, or minimum throughput. There are no existing techniques capable of delivering end-to-end QoS guarantees. This project will develop credit-based controls, instead of the usual rate-based controls, that can deliver these required QoS guarantees. These QoS guarantees will enable the development and deployment of entirely new services on the Internet. This project will give Australian industry the opportunity to create the next generation Internet. Therefore this project represents a Frontier Technology.Read moreRead less
Overseeing the internet: new paradigms of network measurement. Like the electricity network, the internet is a core infrastructure, and so must be reliable and efficient. A gap in bandwidth supply is like a blackout in terms of lost business and productivity. This project will provide the measurement breakthroughs to ensure that network behaviour can be accurately and comprehensively monitored.
Modelling, Measurement and Control of the Internet. The demands made of modern high-speed data networks, such as those that make up the Internet, are increasing rapidly in terms of both the quantity and quality of service required.
So that these demands can be satisfied, it is necessary that network operators, or even the network itself, take measurements and make control decisions. Depending on the service, there is a need to do this over very short timescales (fractions of a second), or lon ....Modelling, Measurement and Control of the Internet. The demands made of modern high-speed data networks, such as those that make up the Internet, are increasing rapidly in terms of both the quantity and quality of service required.
So that these demands can be satisfied, it is necessary that network operators, or even the network itself, take measurements and make control decisions. Depending on the service, there is a need to do this over very short timescales (fractions of a second), or longer
timescales (minutes or even weeks).
By developing innovative mathematical models of network behaviour,
this project will deliver improved methods for measurement and
control of the Internet.Read moreRead less
Increasing internet energy and cost efficiency by improving higher-layer protocols. Australians rely heavily on our telecommunications infrastructure due to our geographic dispersion. We are also very susceptible to climate change, given our reliance on agriculture. Information technology is consuming a rapidly increasing fraction of our power and our budget. This research will help to reverse both those trends, by finding novel and practical ways to use our infrastructure more efficiently, and ....Increasing internet energy and cost efficiency by improving higher-layer protocols. Australians rely heavily on our telecommunications infrastructure due to our geographic dispersion. We are also very susceptible to climate change, given our reliance on agriculture. Information technology is consuming a rapidly increasing fraction of our power and our budget. This research will help to reverse both those trends, by finding novel and practical ways to use our infrastructure more efficiently, and to minimise its energy use. This will enable the Australian telecommunications industry to provide better service (including to Australian industries and rural communities) at lower economic and environmental cost. This project will put Australia on the international stage as a leading contributor to energy-efficient internet technology.Read moreRead less
Modelling Virtual Transport Channel for Future Multimedia Heterogeneous Networks. Recent advances in network bandwidth and processing power have led to the emergence of real time multimedia streaming applications and it is expected that such applications will become an even more dominant source of traffic in future computer networks. Quality of service studies for real-time applications in high-speed networks, point to the importance of the transport layer functionality in admission control, mon ....Modelling Virtual Transport Channel for Future Multimedia Heterogeneous Networks. Recent advances in network bandwidth and processing power have led to the emergence of real time multimedia streaming applications and it is expected that such applications will become an even more dominant source of traffic in future computer networks. Quality of service studies for real-time applications in high-speed networks, point to the importance of the transport layer functionality in admission control, monitoring end-to-end delays, and network bandwidth adaptability. The project will study, design, develop and analyse a comprehensive model for the virtual transport layer to satisfy the necessary quality of service requirements for the next generation of multimedia-based applications.Read moreRead less
Accurate Performance Modelling and Prediction of Cluster Computers. The tools, methodologies and data produced by this project will assist
Australian academic and industrial organisations in choosing the most
cost-effective cluster configurations for their specific high
performance computing requirements. It will also help an Australian
company to compete with increasing strength against the major
multinationals. The project will also draw together and promote future
research links between ....Accurate Performance Modelling and Prediction of Cluster Computers. The tools, methodologies and data produced by this project will assist
Australian academic and industrial organisations in choosing the most
cost-effective cluster configurations for their specific high
performance computing requirements. It will also help an Australian
company to compete with increasing strength against the major
multinationals. The project will also draw together and promote future
research links between two major academic institutions in this field.
Finally, the project will provide high-level training in research,
with industrial grounding, in the high performance computing industry.
Read moreRead less
Providing End-to-End Quality of Service to Internet Applications. Current Internet only supports limited quality of service (QoS) within isolated domains. This project allows critical applications to be deployed over the global Internet. The project employs a programmable architecture for composing services and intelligent software elements for negotiating acceptable QoS responses for individual application flows over multiple administrative domains. The architecture employs DiffServ's aggregati ....Providing End-to-End Quality of Service to Internet Applications. Current Internet only supports limited quality of service (QoS) within isolated domains. This project allows critical applications to be deployed over the global Internet. The project employs a programmable architecture for composing services and intelligent software elements for negotiating acceptable QoS responses for individual application flows over multiple administrative domains. The architecture employs DiffServ's aggregation property to make it scalable for the Internet core. Successful outcomes would open the floodgates, allowing the development and deployment of applications such as high quality videoconferences, live sport broadcasts, remote medical services, network games, real-time on-line educations, and multimedia distributions.
Read moreRead less
Fair Intelligent Quality of Service Control and Path Discovery Mechanisms for Internet Scalable Architectures. Until now the Internet only provides best effort service without guaranteeing the timeliness or actual delivery of data. Quality of Service (QoS) architectures have been proposed but not deployed due to a fundamental problem: either the architecture is not scalable or it does not guarantee end-to-end QoS. The project tackles the problem by introducing mechanisms for network path discove ....Fair Intelligent Quality of Service Control and Path Discovery Mechanisms for Internet Scalable Architectures. Until now the Internet only provides best effort service without guaranteeing the timeliness or actual delivery of data. Quality of Service (QoS) architectures have been proposed but not deployed due to a fundamental problem: either the architecture is not scalable or it does not guarantee end-to-end QoS. The project tackles the problem by introducing mechanisms for network path discovery in the network core. The solution allows an integration of both scalability and end-to-end QoS architectures. The outcomes allow network providers to provision network resources efficiently, to develop competitive Internet-based products; and the service providers to develop quality services such as high quality IP phone, multimedia distributions.
Read moreRead less