J.H. Kim, "Bandwidth and latency guarantees in low-cost, high-performance networks," Ph. D. Thesis, Department of Computer Sciences, University of Illinois at Urbana-Champaign, 1997. Home/Search Document Details and Download
Summary Related Articles Check |
This paper is cited in the following contexts:
On the Buffer Size Requirements of the Multimedia Router (MMR) - Caminero, Carrion, al. (Correct)
....4.2 Scheduling Algorithm: Priority Biasing At the core of the MMR, there is one key element to provide QoS guarantees to the multimedia flows: the link and switch scheduling algorithm. The authors have proposed a link and switch scheduling algorithm, based on the concept of biased priorities [4, 9, 15] that is well suited to parallelization and pipelining [17] The key point in this scheme is that priorities are biased according to the ratio between the QoS a flit is receiving and the one it should receive. This approach combines the effect of the scheduler (measured as the delay or the jitter ....
....the crossbar. At the same time, the link switch scheduling algorithm starts a new execution. See [17] for more details. 4. 3 The SIABP (Simple IABP) Biasing Function In the first switch scheduling proposal for the MMR [17] link scheduling was performed by using the concept of biased priorities [4, 9, 15]. The priority value for each flit was related to the QoS requested by the connection, and was biased depending on the QoS received by it. One of the proposed biasing functions was IABP (Inter Arrival Based Priority) With IABP, the priority value was computed as the ratio between the queuing ....
J.H. Kim, "Bandwidth and latency guarantees in low-cost, high-performance networks," Ph. D. Thesis, Department of Computer Sciences, University of Illinois at Urbana-Champaign, 1997.
QUIC: A Quality of Service Network Interface Layer .. - West.. (1999) (1 citation) (Correct)
....updated at run time to enable QoS requirements to be met. Essentially the packet priority is scaled as a function of the QoS that a packet has experienced up to that point in time relative to the QoS requested by the packet. Such an update operation has been referred to as priority biasing [7, 9, 8] since the priority value is biased by the relative degradation of its service. The biasing operation couples the effect of the scheduler (e.g. queuing delay) with the QoS demand (e.g. jitter bound) This distinguishes this approach from priority update mechanisms such as age counters that do ....
J. H. Kim. Bandwidth and latency guarantees in low-cost high performance networks. Ph.D. Thesis, Department of Computer Sciences, University of Illinois, Urbana-Champaign, January 1997.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC