Huawei Technologies Co., Limited

Tsinghua University

Hao Gong

Guangyu Shi

Jian Ye Chen

Lingyuan Fan

Hao Wu

Jian Chen

Hongliang Yu

Yuyang Du

According to embodiments of the present invention, in a storage system, a switch receives an operation request, where the operation request carries request identification; the switch queries, according to the request identification, a flow table of the switch to obtain a flow table entry matching the operation request; and the switch forwards the operation request to a target storage device among storage devices according to a forwarding rule of the matching flow table entry. By using a characteristic of a SDN, the operation request is processed according to the forwarding rule of the flow table entry of the switch, thereby implementing load balancing of the storage devices and reducing information interaction between the storage devices.

Method and apparatus for processing operation request in storage system

Embodiments of the present invention provide a wear-leveling method, a storage device, and an information system, where a storage region is divided into a plurality of storage sub-regions of the same size. The method includes: recording the accumulated number of write operations of each storage sub-region; and when the accumulated number of write operations of any one storage sub-region of the plurality of storage sub-regions reaches a predetermined remapping rate, mapping a logical address of the storage sub-region to a remapping physical address. With the wear-leveling method, the storage device, and the information system in the embodiments of the present invention, a logical address of a local data block on which too many write operations are performed may be evenly mapped to an overall physical storage region, thereby avoiding that local data is too hot and prolonging a service life of a storage medium.

Wear-leveling method, storage device, and information system

A method and apparatus for maintaining routing information in a Peer-to-Peer (P2P) network are provided. The method includes that: a super maintenance node receives routing update information of an ordinary node in its home area; the super maintenance node sends routing update information to super maintenance nodes in other network areas; the super maintenance nodes in other network areas notify ordinary nodes in their home network areas to update routing information. With the present invention, a node of strong processing capabilities at the network area border is chosen to act as a super maintenance node and other nodes as ordinary nodes; when a node joins or fails, an ordinary node sends detected routing update information only to the super maintenance node in the local network area; the super maintenance node then forwards the update information to super maintenance nodes in other network areas. Each super maintenance node notifies ordinary nodes in its home network area of the received routing update information.

Method and apparatus for maintaining routing information

The disclosure relates to distributed network communications, and in particular, to a distributed network construction method and apparatus, a distributed data storage method and apparatus, and a distributed network system. When a node joins a distributed network, the ID of the node is determined according to the geographic location information about the node. Therefore, all the nodes in the same area belong to the same ID range, and the node IDs are allocated according to the area. Because the node IDs are determined according to the area, the local data may be stored in the node in the area according to the geographic information, inter-area data access is reduced. Therefore, the method, the apparatus, and the system provided herein reduce the data load on the backbone network, balance the data traffic and the bandwidth overhead of the entire network, and reduce the network instability.

Distributed network construction and storage method, apparatus and system

The embodiment of the present invention provides a method for establishing a route, which includes searching a target node by a node from a first level node group according to first level route information, returning to query a result or data saved by the target node if the target node is found, otherwise executing the following step; searching a target node by the node from a higher level node group according to higher level route information, returning to query a result or the data saved by the target node if the target node is found, otherwise repeating the step. Through the hierarchical communication system, most of the flows of daily service operations in the communication network are limited within a small region, thereby preventing from excessively occupying the band width of a backbone network being originally not wide enough.

Method and device for establishing route

A method for predicting a residual online time of a peer in a peer-to-peer (P2P) network is provided. The method includes: obtaining M history life cycle sampling data Sof a peer, i=1, . . . , M; determining the number, n, of Gaussian components in a multidimensional Gaussian Mixture Model to be established, where n&#x2267;2, and the multidimensional Gaussian Mixture Model denotes a probability distribution of a residual online time of the peer; using Sand n to establish the multidimensional Gaussian Mixture Model; and using the established multidimensional Gaussian Mixture Model to predict the residual online time of the peer. A device for predicting a residual online time of a peer in a P2P network is also provided.

Method and device for predicting residual online time of peer in peer-to-peer network

A method for updating routing table information after a node fails in a P2P network is disclosed herein. The method includes: the range of the nodes whose route is directed to the failed node is determined according to the distance between the failed node and the neighboring node of the failed node; the failure information of the failed node is sent to the LDN of the failed node within the node range; and the LDN updates the routing table according to the failure information. A network device and a P2P communication network are also disclosed herein. The embodiments of the present invention enhance the capability of the whole P2P network in perceiving node churn and improve the route search efficiency and system stability of the whole P2P network.

Method, device and system for updating routes after node fails in P2P network

A user data server system includes: a data storage node, which stores user data, registers the user data in a Distributed Hash Table (DHT) network by using a key, and receives and processes user data operation requests; a DHT index node, which creates and maintains DHT routing information according to a DHT algorithm and stores information of data storage nodes where user data is stored according to the key and searches for information of a data storage node where target user data is stored according to the key; a DHT super maintenance node, which manages and optimizes the DHT network; and a front end node capable of protocol processing and service processing, which obtains a key associated with a target user and obtains information of a data storage node where the target user data is stored by querying the DHT index node via the DHT network according to the key, and performs operations on the user data stored in the data storage node where the target user data is stored. With the technical solution provided by the present invention, a user data server is not centralized and is highly scalable and highly reliable with a high cost-effectiveness.

User data server system, method and apparatus

Method and device for establishing a route

In order to overcome the defects of the centralized subscriber data management in the telecommunication network, a highly effective distributed data management system for the communication network is achieved in the present invention through improving and applying the distributed Hash table (DHT) technology to subscriber data management in the communication network, which overcomes the deficiencies of the conventional centralized data management. A method for processing subscriber data in a communication network based on a DHT is provided, which includes the following steps. A first node in a DHT overlay network receives subscriber data to be stored. The subscriber data to be stored is regarded as backup data divided into L/K portions. L is greater than or equal to a ratio of a node working bandwidth of a second node to a node recovering bandwidth for data recovery of the second node in the DHT overlay network. K is a total number of backup data and is an integer greater than zero. The subscriber data to be stored is regarded as master data, and the master data and K&#xd7;L/K divided backup data are, respectively, stored in 1+K&#xd7;L/K DHT nodes.

Growing community of inventors

Shenzhen, China

Hao Gong