Random early detection gateways for congestion avoidance
IEEE/ACM Transactions on Networking (TON)
A Dynamic Priority Assignment Technique for Streams with (m, k)-Firm Deadlines
IEEE Transactions on Computers
Window-Constrained Real-Time Periodic Task Scheduling
RTSS '01 Proceedings of the 22nd IEEE Real-Time Systems Symposium
Dynamic class-based queue management for scalable media servers
Journal of Systems and Software
Dynamic Window-Constrained Scheduling of Real-Time Streams in Media Servers
IEEE Transactions on Computers
Loss-Tolerant QoS using Firm Constraints in Guaranteed Rate Networks
RTAS '04 Proceedings of the 10th IEEE Real-Time and Embedded Technology and Applications Symposium
A Virtual Deadline Scheduler for Window-Constrained Service Guarantees
RTSS '04 Proceedings of the 25th IEEE International Real-Time Systems Symposium
DLB: A Novel Real-time QoS Control Mechanism for Multimedia Transmission
AINA '06 Proceedings of the 20th International Conference on Advanced Information Networking and Applications - Volume 01
Network calculus: a theory of deterministic queuing systems for the internet
Network calculus: a theory of deterministic queuing systems for the internet
Enhanced fixed-priority scheduling with (m,k)-firm guarantee
RTSS'10 Proceedings of the 21st IEEE conference on Real-time systems symposium
Using differentiated services to support Internet telephony
Computer Communications
Hi-index | 0.00 |
This paper presents a new real-time Quality of Service (QoS) scheme, called Relaxed (m, k)-firm, for multimedia flow transmission which can tolerate occasional packet drops. In contrast to traditional real-time constraint, Relaxed (m, k)-firm requires the transmission delay constraint on a group of any k consecutive packets instead of imposing a transmission delay constraint per packet. Due to this relaxation, this novel scheme can achieve high resource utilisation. A new traffic control mechanism, called Double-Leaks Bucket (DLB), is also proposed to selectively drop a proportion of packets in case of network congestion while still guaranteeing the R-(m, k)-firm constraint.