40 results about "Bandwidth cap" patented technology

Bandwidth capping is implemented at a logical port level for a shared Ethernet port. When a physical port of a Host Ethernet Adapter (HEA) is partitioned, a Logical HEA is created for the partition. One or a plurality of Logical Ports (LPorts) is created in the Logical HEA. Each LPort is mapped to a corresponding physical port. During LPAR configuration, a minimum guaranteed speed is specified forthe LPort together with the corresponding physical port for the LPort and an optional maximum speed. The specified configuration for the LPort is verified, and the configuration values are stored inthe HEA and the HEA dispatches data packets based upon the stored configuration values for the LPort.

The invention discloses a traffic routing method, a router and a communication system. In the embodiment of the invention, a dynamic traffic information base is established by collecting data traffic information of links in a network, thus determining the load condition of a current link; and then routing is made according to the load condition of the current link, that is, part of the data traffic of the current link is switched to a light-loaded standby link which is currently invalid to perform transmission if the load is judged to exceed a maximum bandwidth upper threshold K of the current link, thus avoiding the problem of packet loss caused by that the current link directly discards the part of the data traffic which exceeds the bandwidth when the links are congested, effectively protecting user data, allocating bandwidth more reasonably and improving the utilization ratio of the bandwidth.

The present invention provides a communication device and method to avoid the problem of UE repeatedly reporting capability information corresponding to carrier combinations it supports, and reduce the waste of reporting signaling resources. The method of the present invention includes: receiving carrier range information from a base station, where the carrier range information includes at least one of a carrier bandwidth upper limit and a carrier number upper limit; according to at least one of the carrier bandwidth upper limit and the carrier number upper limit, sending capability information to the base station.

The present invention relates to a multi-tenant bandwidth guarantee framework and cost optimization method under a cloud platform, the steps of which are as follows: a tenant scheduler in a pre-set cache framework structure determines a user combination scheme according to user historical access characteristics; The two operations manage the user cache separately to achieve optimal cache space allocation, and control the actual bandwidth obtained by the user based on the equivalent relationship between the bandwidth obtained by the user from the underlying storage and the bandwidth for updating the cache, and strictly limit the upper limit of the bandwidth obtained by the user to the target value The following; the cost of the storage system is divided into the cost of distributed storage and the cost of SSD cache. By adjusting the size of the user cache, the sum of the user's cache cost and the storage layer cost is the lowest, and the system cost is minimized. The invention reasonably allocates the size of the user cache space to maximize the benefits brought by the locality of data access and reduce the internal cost of the system.

The invention belongs to the technical field of visible light communication, and especially relates to a carrier-free amplitude phase modulation system and modulation method based on a Faster-than-Nyquist precoding technology. The system and the method are applied to a visible light communication system. The modulation system mainly comprises a QAM mapping module, an I / Q separation module, a differential coding module, a delay addition module, an upper conversion module and a forming filtering module. When signals are processed, the signals are divided into two ways of same direction I and orthogonal Q branches. The upper way and the lower way are processed by the above-mentioned process. Finally, the two ways of signals are combined into one way of signals at a transmitting end through subtraction operation. Compared with a traditional Nyquist bandwidth CAP modulation system, the system and the method have the advantages that Faster-than-Nyquist CAP signals of which signal bandwidth is further compressed can be generated; the guard band between CAP sub-bands can be eliminated; a certain extent of overlap can be tolerated; the signal bandwidth is utilized effectively; and the spectrum efficiency of the visible light communication system is improved.

The embodiment of the invention discloses a data transmission method and a device of a distributed storage system. The number of connections amonde nodes is calculated according to the bandwidth upper limit of network equipment in advance. The connections of the above number are established among the nodes of the distributed storage system. Meanwhile, a sending queue of a data request and a response queue of the data request are established for each connection. According to the ID values of files / objects in the data request of a sending end, each file / object is placed in the sending queues of different connections in a polling mode, so that the concurrent sending of files / objects is carried out. Meanwhile, the corresponding information of the connections of the files / objects is recorded. When a receiving end processes the data request according to the service logic, the responding information of each file or object in the data request is placed in corresponding response queues according to the corresponding information. Meanwhile, a feedback is concurrently sent to the sending end. Therefore, high-speed network bandwidth resources are fully and effectively utilized, and the service throughput is improved. The request processing time delay is reduced, and the overall performance of the distributed storage system is improved.

The disclosure discloses a method, device and equipment for controlling bandwidth, and belongs to the technical field of computing. The method includes: when the network device provides business services for registered clients, the upper limit of the bandwidth of the network device can be determined first, and then according to the working state of each client and the upper limit of the bandwidth of the network device, Refresh the bandwidth upper limit value of each client, and then reallocate bandwidth to each client according to the bandwidth upper limit value of each client. By adopting the present disclosure, the flexibility of bandwidth allocation can be improved.