A method and device for processing a packet are provided in this disclosure. According to an example of the method, an HTTPS packet is received from a user host, and a non-online user session entry matching the HTTPs packet is searched for according to a source IP address and a destination IP address of the HTTPS packet. In case that the non-online user session entry is found, a token is obtained from a first token bucket if determining that a user session corresponding to the non-online user session entry has no token, where the number of tokens in the first token bucket is set based on processing capability of a CPU of the access gateway device. When the token is successfully obtained, the HTTPS packet is sent to the CPU for processing. When the token has failed to be obtained, the HTTPS packet is abandoned.

A packet queueing system includes an ingress port configured to receive packets; queueing logic communicatively coupled to one or more egress queues for transmission via an egress port, wherein the queueing logic is configured to maintain an Acceptable Burst Size (ABS) token bucket which is set to enable absorption of microbursts, and implement a congestion avoidance algorithm to one of randomly drop packets and queue packets, wherein the congestion avoidance algorithm only performs the randomly drop packets responsive to the ABS token bucket being empty.

An aspect includes splitting a table of buckets into a fixed number of domains. Each of the domains includes a corresponding subset of the buckets. An aspect also includes providing a spare bucket for each of the subsets of the buckets and providing a metadata structure for each of the domains. The metadata structure includes a head pointer that points to a first bucket of a corresponding subset of the buckets and a spare_bucket pointer that points to the spare bucket of the subset of the buckets. An aspect further includes providing a split-spare bucket pointer that interleaves, during updates to data, among the subset of buckets in the domain. Data subject to the updates is stored in the spare bucket for a corresponding one of the domains. An aspect also includes updating the head pointer and the spare_bucket pointer for corresponding domains in response to updating the data.

Methods and systems for dynamic congestion management in communications networks that advantageously provides a satisfactory Quality of Experience (QoE) of real time communication for network users. Congestion management is achieved wherein an ingress interface is monitored by a data processing system and when utilization of that interface exceeds a first activation level a message is sent to a second data processing system wherein that second data processing system is a source for at least some of data packets traversing the ingress interface, wherein the first message indicates that traffic shaping is to occur in accordance with the first activation level and only if the utilization falls below a deactivation level, transmitting a second message to the second data processing system wherein the second message indicates that traffic shaping is to stop.

This application provides a resource allocation method and a communications device. The method includes: receiving, by a communications device, logical channel configuration information sent by a network devicetime duration; and determining, by the communications device based on the logical channel configuration information, whether to allocate a resource to a data packet on the logical channel. According to the resource allocation method provided in this application, when needing to allocate a transmission resource to the data packet on the logical channel, the communications device may determine, based on the time duration of the logical channel and a data volume of data that can be transmitted on the logical channel within the time duration that are in the logical channel configuration information sent by the network device, whether to allocate the resource to the data packet on the logical channel.

Various embodiments provide a resource allocation method and an apparatus. In those embodiments, a terminal device obtains a first parameter indicating a maximum service data volume to be provided by an access network device for a first service in a first time length; and determines, based on the maximum service data volume, a resource of a media access control protocol data unit MAC PDU, the resource being occupied by buffered data of the first service. In those embodiments, the terminal device determines, based on the maximum service data volume of the first service, the resource of the MAC PDU. Compared with a conventional method, the maximum service data volume of the first service is considered. This helps improve reasonableness of allocating the resource of the MAC PDU to the data of the first service.

A method and a network node for traffic shaping of a packet switched network is presented, the network node being arranged for processing packets to be transmitted in the network according to at least: a serial packet processing algorithm A.sub.ser providing a synchronized utilization of a set of at least one processing unit; anda parallel packet processing algorithm A.sub.par providing an at least partly unsynchronized utilization of the set of at least one processing unit; whereinthe processing of the packets corresponds to a total packet cost, which is cooperatively shared by the at least one processing unit of the set. The method further includes: determining, when the processing according to the parallel packet processing algorithm A par is used for processing the packets, if shares of the total packet cost for one or more of the at least one processing units exceed a capacity to process packets for the at least one processing units, respectively, wherein each one of the shares corresponds to one or more packet; andswitching from the processing according to the parallel packet processing algorithm A.sub.par to processing according to the serial packet processing algorithm A.sub.ser if the shares of the total packet cost for one or more of the at least one processing units, respectively, is determined to exceed the capacity.

In various embodiments, methods and systems for implementing network traffic flow logging in a distributed computing system are provided. At a high level, the network traffic flow logger is implemented using a network traffic flow logger framework that includes a plurality of modular network traffic flow logging objects for adjustable processing of network flow events. In operation, a plurality of tokens is counted at a first throttling rate. The first throttling rate is associated with a maximum number of tokens threshold. The maximum number of tokens threshold is an adjustable threshold. A flow event is accessed, where flow events are quantified based on tokens. The flow event is queued, based on a count of the plurality of tokens or the maximum number of tokens threshold. Based on queueing the flow event, the first throttling rate is adjusted to a second throttling rate based on a queue length of the queue.

Systems and methods for rate limiting one or more clusters of service instances using at least one rate limit controller are described herein. A token distribution is determined for each one of a plurality of rate limiters. The token distribution comprising a maximum number of tokens and a token generating rate. The maximum number of tokens and the token generating rate are assigned to each one of the plurality of rate limiters. At least one request for additional tokens is received from at least a given one of the plurality of rate limiters. The token distribution of at least the given one of the plurality of rate limiters is adjusted based on the request and on token consumption information of at least the given one of the plurality of rate limiters. An adjusted token distribution is assigned to the given one of the plurality of rate limiters.

A system to support Time Division Duplex (TDD) Full Duplex (FDX) with a remote Duty Cycle (DS) framer may be provided. The system may comprise a core and a Remote PHY Device (RPD). The core may comprise a Downstream (DS) Quality-of-Service (QoS) scheduler and a DS duty cycle rate regulator. The DS duty cycle rate regulator may comprise a plurality of token buckets and a plurality of data framing queues. The RPD may comprise a plurality of TG queue blocks and switching logic.