Fast packet switching

About: Fast packet switching is a research topic. Over the lifetime, 5641 publications have been published within this topic receiving 111603 citations.

Method And Apparatus For Generating Packet Frames For Carrying Data

Abstract: A conventional packet frame includes a MAC header, one MAC service data unit (MSDU) and a frame FCS. This frame is repeated to send a plurality of MSDUs. A proposed packet frame includes a MAC header, a plurality of MSDUs and a frame FCS. A unique bit pattern is provided in the MAC header to distinguish the proposed packet frame from the conventional packet frame. Thus, the total data stream can be shortened.

Flow based flow control in an ethernet switch backplane

Abstract: In one embodiment, a method includes detecting packet congestion in a network device that includes a switch, and a plurality of components operable to transmit and receive data packets from and to the switch. The method includes, in response to detecting packet congestion, transmitting an Ethernet pause frame to at least one of the plurality of components, the at least one component identified as a source of a data packet that caused the packet congestion, and transmitting a backward congestion notification (BCN) packet to the at least one component, the BCN packet including some data payload of the data packet that caused the packet congestion. The method includes regulating an offending packet stream originated from the at least one component relative to a non-off-ending packet stream originated from the at least one component to lower traffic load of the offending packet stream relative to the non-offending packet stream, where the data packet that caused the packet congestion belonging to the offending packet stream.

Performance analysis of data packet discarding in ATM networks

Abstract: Data performance in ATM networks should be measured on the packet level instead of the cell level, since one or more cell losses within each packet is equivalent to the loss of the packet itself. Two packet-level control schemes, packet tail discarding and early packet discarding, were proposed to improve data performance. In this paper, a new stochastic modeling technique is developed for performance evaluation of two existing packet-discarding schemes at a single bottleneck node. We assume that the data arrival process is independent of the nodal congestion, which may represent the unspecified bit-rate traffic class in ATM networks, where no end-to-end feedback control mechanism is implemented. Through numerical study, we explore the effects of buffer capacity, control threshold, packet size, source access rate, underlying high-priority real-time traffic, and loading factor on data performance, and discuss their design tradeoffs. Our study shows that a network system can he entirely shut down in an overload period if no packet-discarding control scheme is implemented, under the assumption that there are no higher layer congestion avoidance schemes. Further, unless with sufficiently large buffer capacity, early packet discarding (EPD) always outperforms packet tail discarding (PTD) significantly under most renditions. Especially under the overload condition, EPD can always achieve about 100% goodput and 0% badput, whereas the PTD performance deteriorates rapidly. Among all the factors, the packet size has a dominant impact on EPD performance. The optimal selection of the EPD queue control threshold to achieve the maximum goodput is found to be relatively insensitive to traffic statistics.

Controlling packet congestion

Abstract: A packet congestion control technique employs an Active Flow Random Early Drop (AFRED) method to provide a service requiring a predetermined bandwidth with a stable Quality of Service (QoS) even when network congestion occurs. The packet congestion control technique includes: receiving at least one packet flow; defining identifiers for the received at least one packet flow; numbering each of the defined at least one packet flow; calculating a queue length of each of the at least one packet flow; allocating a maximum queue threshold value, a minimum queue threshold value, and a maximum probability value of each of the at least one packet flow; and dropping or accepting an incoming packet of a flow according to the calculated packet drop probability upon a queue length of a currently received packet being between the maximum queue threshold value and the minimum queue threshold value.

Field-trial demonstration of cost efficient sub-wavelength service through integrated packet/circuit hybrid network [invited]

Abstract: Carriers are under constant pressure to meet the ever-increasing bandwidth demand while reducing cost per bit, enhancing network throughput, and offering a large variety of services. Hybrid packet and circuit network technologies are being widely investigated and considered as a solution for offering both the high network throughput of the packet domain and wavelength services, i.e., a low fixed latency and zero packet loss. To enable carriers to serve a higher number of customers requiring wavelength services, optical transport network (OTN)-based sub-wavelength switching is adopted to support finer granularity with similar performance to full wavelength services. However, OTN is not able to perform statistical multiplexing and achieve the throughput efficiency of packet networks. In this work an integrated hybrid optical network field-trial is described to demonstrate the ability to both aggregate and transport sub-wavelength circuits, and offer high throughput efficiency by statistically multiplexing traffic on transport wavelengths. Results show the transport of sub-wavelength services with packet-delay variation limited to only 15 ns and 82.4% wavelength utilization using statistical multiplexing.