End-to-end packet delay and loss behavior in the internet
SIGCOMM '93 Conference proceedings on Communications architectures, protocols and applications
Driving resource management with application-level quality of service specifications
Proceedings of the first international conference on Information and computation economies
QoS routing in networks with inaccurate information: theory and algorithms
IEEE/ACM Transactions on Networking (TON)
Providing guaranteed services without per flow management
Proceedings of the conference on Applications, technologies, architectures, and protocols for computer communication
Endpoint admission control: architectural issues and performance
Proceedings of the conference on Applications, Technologies, Architectures, and Protocols for Computer Communication
QUIC: A Quality of Service Network Interface Layer for Communication in NOWs
HCW '99 Proceedings of the Eighth Heterogeneous Computing Workshop
Statistical multiplexing using MPEG-2 video encoders
IBM Journal of Research and Development
Web-based intranet services and network management
IEEE Communications Magazine
IEEE Journal on Selected Areas in Communications
Statistical multiplexing of multiple time-scale Markov streams
IEEE Journal on Selected Areas in Communications
Pricing congestible network resources
IEEE Journal on Selected Areas in Communications
| Hi-index | 0.24 |
The paper describes a management-oriented model for cost-effective 'data connectivity' provisioning between the peer entities of a networked multimedia application. The service provider (SP) may employ policy functions to map the application-prescribed flow specs onto the resource needs of connections carrying data flows. The model allows dynamic switching from one policy function to another, based on a notion of cost associated with the infrastructure resource usage, namely, the network bandwidth, for a given level of connectivity provisioning. Our strategy is to reduce the per-flow cost incurred by multiplexing many closely similar data flows on a single connection. The multiplexing brings in two benefits to the SP, without compromising the QoS needs of applications. First, it reduces the per-flow resource allocation due to the gains accrued from a statistical sharing of connection resources. Second, it amortizes the connection-level overhead across many flows. The level of cost reduction, and hence the revenue accrual, can be controlled by the SP using a range of policy functions that take into account the burstiness and loss/delay tolerance of data flows. The paper describes the functional mechanisms and signaling structures to realize our model of connectivity services. We have conducted studies of different types QoS-controlled data connections based on traffic measurements and simulation analysis to validate the model. The model can be easily incorporated into the TMN and DCOM frameworks that have been standardized for network service offerings.