46results about How to "Guaranteed Bandwidth Requirements" patented technology

The invention discloses a routing algorithm of guaranteeing QoS (quality of service) in a data center network based on SDN. Aiming at problems that user terminal transmission time delay is high and a link congestion is generated in the data center network, the routing algorithm of guaranteeing QoS in the data center network is provided. In the algorithm, priority classification is performed on a data flow arriving at a switch so that a high priority data flow which is sensitive to the time delay adopts a time delay guarantee algorithm; a path with minimum data plane time delay is found to carry out flow table issuing; for a low priority data flow which has a high requirement to a bandwidth, a bandwidth guarantee algorithm is adopted; and according to a demand of the data flow for the bandwidth, dynamic bandwidth adjusting is performed. The QoS of a user is effectively guaranteed and a resource utilization rate in the data center network is increased.

The invention relates to the field of the mobile Internet and provides wireless access equipment for user network bandwidth dynamic allocation. At least two groups of users are allowed to be connected in. The wireless access equipment is characterized in that the wireless access equipment detects data flow input by access users, and according to the detecting result, the network bandwidths of the users are adjusted dynamically. When the speed of the data flow input by the access users in one or more groups of users is increased, the bandwidth which is allocated for external sharing users is reduced. When the speed of the data flow input by the access users in one or more groups of users is reduced, the bandwidth which is allocated for the external sharing users is increased. According to the needs for different services at different time periods of inner users, the bandwidth allocation of the inner users and the external sharing users is adjusted dynamically, the fact that the requirements for the bandwidths of the inner users are met can be guaranteed, and meanwhile surplus bandwidth resources are provided for the external users to share.

The invention discloses a passive optical network bandwidth allocation method. The passive optical network bandwidth allocation method comprises the steps that after an optical link terminal (OLT) receives a request bandwidth allocation data package sent by a center optical network unit (CONU), bandwidth allocation is carried out according to a preset dynamic bandwidth allocation (DBA) rule; the OLT obtains a first logical link identifier (LLID1) and a second logical link identifier (LLID2) in the request bandwidth allocation data package according to the request bandwidth allocation data package; the OLT packages a bandwidth allocation result in a bandwidth allocation result data package through the LLID1 and the LLID2; the OLT sends the bandwidth allocation result data package to the CONU. The method is applicable to the long-distance PON, the DBA is finished through two levels of control mechanisms in the mode that ONUs are grouped, the ONUs adopt the LLID1 and the LLID2 and an intelligent ODU is utilized, the bandwidth requirement of users can be met, the bandwidth utilization rate is improved, and the time delay is reduced.

The invention relates to a WiMAX downlink dispatching method which is characterized by comprising the following five steps sequentially of: (1) generating a 'momentum' parameter according to different time-delay requirements of each business flow and different channel situations of each user; (2) storing the generated 'momentum' parameter in a 'momentum' parameter buffer memory in BS; (3) ranking the 'momentum' parameter in the 'momentum' parameter buffer memory according to a descending order, and dispatching the corresponding business flow according to the ranking order, thus forming a WiMAX frame; (4) carrying out 'momentum' updating, and storing the updated 'momentum' parameter in a 'momentum' buffer memory together with the newly formed 'momentum' parameter; and (5) circulating the steps (1) to (4), thus forming a new WiMAX frame. The method not only can meet the time-delay requirement of real-time services, but also can consider the system throughput simultaneously, furthermore, the calculation is simple, and the computation workload is light.

The invention provides a transfer, calculation and storage integrated cooperation system and method. The system architecture comprises a transfer, calculation and storage service sensor, a transfer, calculation and storage service recognizer, a service strategy and decision processor, a transfer, calculation and storage service processor, and a network management control module, wherein the network management control module adopts a network control technology based on an SDN, and the accuracy of paths and time delays is realized. According to the transfer, calculation and storage integrated cooperation system and the method, the mechanism problem of service perception, scheduling and distribution in calculation, storage and network integrated collaboration can be solved, the system scheme mainly comprises the transfer, calculation and storage service sensor, the transfer, calculation and storage service recognizer, the service strategy and decision processor, the transfer, calculation and storage service processor and the like, and the network, calculation and content are perceived, processed and dispatched, so that efficient service distribution is achieved, and meanwhile the utilization efficiency of calculation, storage and network resources is improved.

A method and apparatus are disclosed for adjusting the bandwidth of a connection between at least two communication endpoints in a data network via a monitoring unit and a control unit. The connection in the data network is assigned at least one transmission channel for data transmission. The user data is, in this case, allocated to at least one communication connection; in particular a voice connection. The monitoring unit monitors the signaling connection for requests for at least one further communication connection; in particular, a voice connection. When a request occurs, it signals to the control unit to assign to the connection one or more additional free transmission channels for the at least one requested communication connection.

The invention provides a virtual resource management method and device, and belongs to the technical field of the virtualization. The method comprises the following steps: determining a target virtualnetwork card from multiple virtual network cards corresponding to a target network card, wherein the target network card is the network card with the residual bandwidth greater than the target bandwidth, and the target virtual network card is the virtual network card not binding with any virtual machine at present; binding the target virtual machine with the target virtual network card, and mapping the target virtual network card to the target virtual machine; acquiring port information for indicating the target network card and the target virtual network card, and returning the port information to a virtualization management system VMS. According to the target bandwidth required by the message transmission, the network card with the residual bandwidth greater than the target bandwidth isused as the target network card, and the target virtual network card is distributed to the target virtual machine in the target network card, the virtual network card capable of performing mapping isactively provided for the user, and the demand on the bandwidth by the user in the message transmission is guaranteed, and the intelligence is high.

The invention provides a long-distance passive optical network protecting method orienting single SRLG faults and belongs to the technical field of communication networks. According to the method, the reliability of each ONU and the union reliability among different ONUs are calculated according to a reliability evaluation model; protected service volume, newly increased and deployed between any pair of ONUs, of each back-up optical fiber is calculated, and then the cost efficiency of the back-up optical fibers is calculated out; the pair of ONUs with the maximum cost efficiency is decided, a back-up optical fiber link is deployed between the pair of ONUs, minimization of the back-up capacity newly increased is taken as the target, and the back-up capacity between the two ONUs is assigned; back-up capacity assignment and remaining back-up optical fiber deployment are carried out in an iterated mode according to the steps until all the ONUs in a network are protected completely; a back-up capacity assignment solution set and a back-up optical fiber deployment position are output. According to the long-distance passive optical network protecting method orienting the single SRLG faults, on the constraint of meeting the requirement for the reliability and protecting service completely, the deployment cost of the back-up optical fibers is minimized.

The invention discloses a virtual network resource allocation method with high reliability and load balance. A vEPC and a network link is included, and the vEPC virtualizes basic hardware equipment into universal network resources; according to the resource allocation method, basic hardware resources in universal network resources are used as an underlying network, the virtual service carried by the underlying network is called as a virtual network; the basic hardware resources in universal network resources are used as the underlying network, the underlying network is divided into communitiesand the communities are optimized; the underlying network and the virtual network are sorted; and finally, resources are allocated to the virtual network. The method comprises two processes of allocating resources to virtual nodes and allocating resources to virtual links, load balance and reliability are guaranteed when the resources are allocated to the virtual nodes, bandwidth requirements ofa virtual network are guaranteed when the resources are allocated to the virtual links, and the problem of large-scale virtual network resource allocation of a 5G core network is effectively solved.

The invention discloses a group switching method based on service priority in blocking rate constraint. According to the method, firstly group switching users in a network are distinguished according to service types, at the same time different priorities are distributed for different types of users, thus a priority queue is formed, and the users are switched orderly according to the priority queue. When the users are in the same priority, a blocking rate function is constructed, and through limiting the current blocking rate in given blocking rate constraint, the number of the users which can be switched at current time slot is determined. After the number of the users which can be switched is determined, only the above users can be switched at present, and thus a network congestion problem can be effectively avoided. The method comprises the following steps: (1) classifying the users and generating the priority queue, (2) determining the probability of networks selected by the users, (3) defining a switching blocking rate, (4) determining the number of the switched users in each time slot, and (5) executing the switching concretely.

The embodiment of the present application discloses a resource allocation method of a hard disk in a distributed storage system. The distributed storage system comprises a plurality of hard disks. Themethod includes selecting a hard disk set from the plurality of hard disks according to a bandwidth requirement and a storage specification of a first user, wherein the storage bandwidth of the firststorage space of each hard disk in the hard disk set satisfies the bandwidth requirement; determining the number of hard disks in the hard disk set by the storage specification, wherein the storage specification is based on a multi-copy or erasure code technology; creating a first logical storage space for the first user from the hard disk set, wherein a first storage space of each hard disk in the hard disk set provides a storage space for the first logical storage space. A low bandwidth limit of a hard disk can be guaranteed by an embodiment of that present application.

An embodiment of the present invention provides a service quality improvement method, the method includes: receiving a first service quality improvement request sent by a first user equipment, where the first service quality improvement request is used to request to improve the first user equipment and the first service quality improvement request. quality of service between two user equipments, the first service quality improvement request carries the identification of the first user equipment and the second user equipment and the service quality type; according to the first service quality improvement request, determine service quality requirements; send the service quality requirements to the policy and charging function entities of the first user equipment and the second user equipment, and the policy and charging function entity policies and charging control rules, After that, the policy and charging function entity sends the policy and charging control rules to the gateways accessed by the first user equipment and the second user equipment, for the first user equipment and the second user The device enhances the quality of service. With the method provided by the embodiment of the present invention, the quality of service between multiple user equipments that are performing wireless services can be improved.

The invention discloses a reactance matching type and distributed type fused ultra-wideband power chip circuit. The respective advantages of a reactance matching type circuit matching structure and adistributed type circuit matching structure are fused; a reactance matching type structure is adopted in a radio frequency power output stage of the ultra-wideband power amplifier circuit, so impedance matching can be easily and accurately output to improve the efficiency; and the power driving stage adopts a distributed structure, so the circuit bandwidth is easy to expand and the gain flatness is easy to improve. The advantages of the two circuit structures are fused, and the performance of the millimeter-wave ultra-wideband high-power monolithic integrated power amplifier can be greatly improved.

Embodiments of the invention provide a service flow aggregation method and device. The method comprises the following steps of computing a preset attribute value and traffic of each service flow for the service flows which reach the same source and the same target within the preset time bucket; dividing the service flows with the similarity of the preset attribute values into one group to obtain a pre-divided service flow group, wherein the similarity of the preset attribute values is that the preset attribute values conform to predetermined similar conditions; and respectively carrying out service flow aggregation on each obtained pre-divided service flow group to obtain a target service flow group which is about to make a request for a forwarding path to a controller. In a service flow aggregation process provided by the embodiments of the invention, the preset attribute values of the service flows in each pre-divided service flow group are similar, so that the preset attribute values of the service flows in the target service flow group are also similar. Meanwhile, the total traffic of the service flows in the target service flow group is not greater than a preset threshold. Therefore, the proper path is subsequently distributed to the target service flow conveniently and effectively, and compared with the prior art, the effectiveness of aggregation is improved.

The invention discloses an I/O service quality maintaining method for a heterogeneous storage cluster system. The method comprises the following steps that firstly, storage resources in a quick storage partition of the heterogeneous storage cluster system are divided into a plurality of storage resource packets, correlation does not exist in the I/O process among the storage resource packets, and the performance of the different storage resource packets cannot be affected during the respective I/O process; secondly, when the heterogeneous storage cluster system receives jobs, a current job is allocated to a storage resource packet selected from the storage resource packets according to the output performance requirement level of I/O of the current job and the states of the storage resource packets, and the states of the storage resource packets are updated. The I/O service quality maintaining method has the advantages that the I/O service performance is high, the I/O bandwidth requirement is low, the I/O service quality is good, and the method is especially suitable for I/O intensive job.