52results about How to "Even load sharing" patented technology

The method includes steps: receiving message flow, and adding sequence number for message in the message flow; distributing message in the message flow to N pieces of processing unit evenly to carry out process (N as natural number larger than 1); based on sequence number to carry out ordering for messages processed by N pieces of processing unit. The invention also discloses a system for processing message. The invention guarantees that messages are not disordered. Load is partaken among multiple PUs evenly. Thus, multiplying processing units realizes multiplying bandwidth of interface.

The present invention discloses a data flow sharing method and equipment. The data flow sharing method is applied for sharing data flow of an equivalent forwarding path of a distributed system. Network equipment is connected with one another through interconnection ports among the equipment. The data flow sharing method comprises the following steps: receiving a counted value of service forwarding port data flow reported by non-control equipment on the equivalent forwarding path by the control equipment in the distributed system, and in combination with the counted value of the service forwarding port on a same equivalent forwarding path, respectively calculating actual total data flow and to-be-allocated total data flow of each network equipment; comparing the actual total data flow and to-be-allocated total data flow of each network equipment by the control equipment, if a fact that the actual total data flow of the network equipment is larger than the to-be-allocated total data flow is found, using the network equipment as shunted equipment, selecting target network equipment for the shunted equipment, and triggering the shunted equipment to forward corresponding data flow to the target network equipment through the interconnection ports between the shunted equipment and the target network equipment according to a difference between the actual total data flow of the shunted equipment and the to-be-allocated data flow.

The embodiment of the invention discloses a load sharing balance method and device for physical ports, and a link aggregation system. The load sharing balance method comprises the following steps: determining that load sharing among physical ports on other positions is uniform through a position of the physical port affecting load sharing in a forwarding table in a link aggregation group through first Hash calculation; then determining the specific position and range of the physical port for forwarding data through secondary Hash calculation, so as to ensure that the physical port which is selected from the determined physical ports in the forwarding table is used for forwarding data in the data forwarding process, thus ensuring load share balance of the flow of each physical port in the link aggregation group added in the data forwarding process. The purposes of data forwarding balance and sufficient utilization of bandwidth of the physical ports are achieved.

The present invention discloses an inner reliable interconnected communication apparatus. The present invention adopts port binding technology to realize communication apparatus reliable interconnecting, through port binding two mutual backed physical network board communications to form one logical network board. In-frame service boards are all bound with tow physical network boards for realizing in-frame communication, logical network among different frames are all interconnected; physical network board in same frame also is bound; on this basic, service board or network board capable of select port communication to realize service load equipartition in port; in port fault automatically abandoning said link to realize link level protection reversion.

The embodiment of the invention provides a load sharing method and routing equipment. All routers with load sharing relationship in a network are used for determining the respective load sharing information by a negotiation method, so as to avoid the condition that the disturbance factor of the respective load sharing information is the same as the disturbance factor of other routing equipment. Specifically, each routing equipment with load sharing relationship is used for obtaining the load sharing information of all downstream routing equipment with the load sharing relationship; when the respective load sharing information is determined, the respective load sharing information is determined according to the load sharing information of all downstream routing equipment with the load sharing relationship, so the respective disturbance factor is different from the disturbance factor of all downstream routing equipment with the load sharing relationship. The routing equipment has the advantages that the coupling degree of the load sharing information of all routing equipment with the load sharing relationship in the network is reduced, and the load sharing at each level is uniform.

The invention discloses a system for realizing the routing function under the multi-gateway model of a core network, which comprises a network management system, at least one MSC Server and at least four MGW. The MSC Server comprises a calling control module, a routing selection module and a storage module; the network management system manages the MSC Server and MGW and carries out the parameter configuration and stores the configuration parameters; the calling control module receives the calling of the gateway at the side of an incoming bureau and controls the MGW on the bearer path to establish the bearing according to the information of the bearer link; the routing selection module calculates the tension degree parameters of different paths according to the total number of the configuration resources and the total number of the occupied resources and selects the bearer path and sends the information to the calling control module; the storage module stores the configuration parameters of the routing resources. The invention also discloses a method for realizing the routing function under the multi-gateway model of the core network and overcomes the defect that the unfair allocation of the resources between gateways causes the congestion and ensures that the entire network runs controllably, steadily and fairly.

The invention provides a message sending method and device. The method and the device can be applied to any member device of any station in a resilient packet ring RPC network. The method comprises the steps of receiving Ethernet unicast massages; searching an RPR MAC (Media Access Control) addresses corresponding to the target addresses of the Ethernet unicast massages; searching corresponding unicast RPR aggregation rings according to the searched RPR MAC addresses; calculating a first ring selection hash value; when the first ring selection hash value matches an RPR logic port in the searched unicast RPR aggregation rings, sending the Ethernet unicast massages to the member devices on an RPR single ring corresponding to the RPR logic port matching the first ring selection hash value, wherein the member devices belongs to a local station; and when the first ring selection hash value matches local RPR single ring outcoming interfaces in the searched unicast RPR aggregation rings, packaging the Ethernet unicast massages into RPR unicast messages according to searched RPR MAC unicast addresses, and sending the RPR unicast messages through local RPR single ring outcoming interfaces matching a second unicast hash value. Message forwarding loads are shared in the RPR network.

The application provides a message broadcasting method and device based on a VXLAN tunnel. According to the application, a tunnel next hop is selected from N equivalent tunnel next hops of the VXLAN tunnel according to a uniform load sharing mode, and thus the process that different broadcast messages are uniformly shared to each tunnel next hop of the VXLAN tunnel can be finally achieved, and theload sharing of ECMPs of the N equivalent tunnel next hops in the VXLAN tunnel can be realized.

The invention discloses a method for selecting a multi-gateway router with central management. The method comprises the following steps: a call control server picks out the paths with the hop numbers which are less than or equal to a collocated hop number between a coming-in side gateway and a going-out side gateway according to the network topological collocation; the modification average transmission line stress degree of each path is respectively calculated according to a total resource collocation number and a occupied resource number; the path with the least modification average transmission line stress degree is selected from all the paths as the bearer path. The invention also discloses another method for selecting a multi-gateway router with central management. In the method, if the stress degree factors of each route of the shortest path are all less than a collocated load start threshold, the shortest path is selected as the bearer path for current call; otherwise, the path with the least modification average transmission line stress degree is the bearer path. The invention solves the problem of congestion caused by unfair distribution of resources among gateways. Thus, the whole network can operate steadily, fairly and controllably.

In a belt/chain drive system, pairs of drive wheels provided with a helical cut spline on the inner diameter of the drive wheels are mounted on a drive shaft provided with helical cut splines on the outer diameter of the drive shaft. The pairs of drive wheels engage via pairs of closed loop belts/chains pairs of driven wheels provided with a helical cut spline on the inner diameter of the driven wheels mounted on a driven shaft provided with helical cut splines on the outer diameter of the driven shaft. In operation, there is equal load sharing among drive wheels and driven wheels.

The invention provides a CGN inter-board load sharing method and device. The method comprises the following steps of receiving a first message copied and sent by a network side service board, whereinthe first message is copied and sent when the network side service board detects that the first IP address of a first message received by a network side interface is the same as the public network IPaddress of a locally configured network address translation service NAT server, and the first IP address is a destination IP address of the first message; if a session table item matched with the first message does not exist locally, carrying out NAT conversion on the first IP address of the first message to obtain a second IP address, wherein the second IP address is the IP address of the user side equipment; determining a target CGN board of the first message according to the second IP address, judging whether the board is the target CGN board or not, and if not, discarding the first message; and if yes, forwarding the first message to the user side service board. Therefore, the load sharing between the CGN boards can be more balanced.

The invention discloses a load balancing pruning method of the loop Ethernet. The method is used for solving the problems that in a traditional loop Ethernet networking pruning method, network load shairing is ununiform, tail-end slave node data transmission delay is long. The method comprises the steps that 1, a master node detects the port connecting state; 2, a slave node sends a link initialization frame; 3, the difference of the distances between the slave node of the loop Ethernet and two ports of the master node of the loop Ethernet is acquired; 4, a redundant link is pruned; 5, a user accesses a network except the loop Ethernet. According to the method, the transmission capacity that the master node blocks the port is fully utilized, and network load balancing is achieved; the node blocking position is reasonably selected, and network delay generated when data of the tail-end slave node is introduced through the loop Ethernet is reduced.