摘要:
There is provided a communication device for controlling an upper limit bandwidth of TCP communication for transmission. There is further provided a communication device for enhancing a communication bandwidth for TCP and restricting it to an upper limit for transmission by use of new TCP for enhancing a communication bandwidth. The device comprises a transmission bandwidth control unit for determining a communication bandwidth of each session of TCP or new TCP, a token bucket update unit for determining whether packets are transmittable per session, and a maximum bandwidth control unit for determining whether packets are transmittable based on an upper limit of a total sum of all the set sessions, wherein when the maximum bandwidth control unit determines that packets are transmittable, the token bucket update unit transmits packets of a session determined as transmittable.
摘要:
In a network communication apparatus, transfer and discard processing is realized for each transfer frame having a priority, and a band limitation is carried out to prevent the congestion of a transfer destination apparatus. Specifically, a bucket is provided for every priority, and tokens of an amount determined based on the priority are supplemented. Moreover, before a bucket evaluation, one shared bucket is provided. All of the tokens existing in the bucket provided for every priority are transferred to the shared bucket for every constant period. The tokens overflowing beyond a capacity of the shared bucket are discarded. All of the frames are equally evaluated and transferred regardless of the priority of each of the frames, as far as the token exists in the shared bucket. When there is no token in the shared bucket, switching to the preferential band limitation is carried out and each priority bucket is evaluated. The bucket corresponding to the priority of the frame is checked and when the token exists, transfer processing of the frame is carried out. When there is no token, discard processing of the frame is carried out.
摘要:
A method of prioritising packets for delivery over a transport network interconnecting nodes of a mobile network, where a guaranteed minimum information rate over the transport network is specified for the mobile network. The method comprises, for each bearer to be injected into the transport network from the mobile network, specifying a bearer information rate and marking packets up to that rate as conformant with the bearer information rate and marking packets exceeding that rate as non-conformant with the bearer information rate. A plurality of traffic type streams from the mobile network are converged, each traffic type stream comprising a plurality of bearers. Packets of the converged traffic type streams are inspected to identify packets marked as conformant and non-conformant, and re-marking non-conformant packets, or at least a fraction of non-conformant packets, as conformant if the converged rate of conformant packets is less than said minimum information rate. The transport network prioritises the delivery of packets marked as conformant over those packets marked as non-conformant.
摘要:
Embodiments of the invention include a method performed in a packet processor core for policing a packet through a hierarchical policer coupled to one or more policing requestors. The hierarchical policer has a plurality of meter levels including an initial level and one or more subsequent levels. The hierarchical policer creates a meter result at the meter of each meter level using packet characteristics and a meter state for that meter level. The hierarchical policer generates meter level outputs that classify the packet for each meter level and for at least one of the subsequent levels the meter level output is based on the meter level output from a previous meter level. The hierarchical policer performs a meter combine operation that produces a final packet output attribute from the combination of the meter level outputs. The hierarchical policer returns the final packet output attribute to a policing requestor.
摘要:
Methods and systems for policing usage by one or more users (12) of a data network (14), said one or more users (12) being able to send, to receive, or to send and receive data over the network via one or more of a plurality of access nodes (42), the usage of the data network resulting from respective data units consuming network resources along respective paths across the data network (14), each of said data units being associated with one of a plurality of subsets of the usage of the data network, said data units having congestion indications associated therewith indicating a measure of their congestion impact.
摘要:
A node (260, 50) for a multi-token optical communications network has optical channels between the node and other nodes, each channel having a token (T1, T2, T3), passed between nodes, to indicate that a corresponding optical channel is available for transmission during a token holding time. The node has a transmitter (280) for transmitting packets over the optical channels, a buffer (170, 270) for queuing packets before transmission, and a transmit controller (170, 290) configured to control the buffer to forward an initial packet or packets from the buffer to the transmitter once a token has been received. The transmit controller determines how much of the token holding time remains after the transmission of the initial packet or packets, and then controls the buffer to forward a further packet according to the remaining token holding time. A maximum packet delay can be reduced where there is asymmetric traffic. A token holding time can be different for different nodes.
摘要:
Computerized methods, systems, and computer-storage media for allowing virtual machines (VMs) residing on a common physical node to fairly share network bandwidth are provided. Restrictions on resource consumption are implemented to ameliorate stressing the network bandwidth or adversely affecting the quality of service (QoS) guaranteed to tenants of the physical node. The restrictions involves providing a scheduler that dynamically controls networking bandwidth allocated to each of the VMs as a function of QoS policies. These QoS policies are enforced by controlling a volume of traffic being sent from the VMs. Controlling traffic includes depositing tokens into token-bucket queues assigned to the VMs, respectively. The tokens are consumed as packets pass through the token-bucket queues. Upon consumption, packets are held until sufficient tokens are reloaded to the token-bucket queues.
摘要:
A token bucket controller for data packet shaping or scheduling, has a bucket parameter of M bits, M being a first integer value. The bucket parameter comprises a bucket value of N bits and a compression parameter of M-N bits for extending the bucket range, N being a second integer value smaller than M.
摘要:
A method of dynamic bandwidth allocation comprising updating a limit value (B max,i,j ) of bandwidth, comprising subtracting from an initial value (N ij ) a value (M ij k ) indicative of previously assigned bandwidth, and adding a predetermined bandwidth value (T ij ).