Unlock instant, AI-driven research and patent intelligence for your innovation.

A Distributed Keyword Table Network and Its Operation Method

A keyword table and keyword technology, applied in the field of computer networks, can solve the problems of node A's unbalanced load, influence, and inability to achieve load balance, and achieve the effect of flexible load distribution, load balance, and low cost

Inactive Publication Date: 2017-08-25
ZTE CORP
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Since the balance of the backup load is limited by the size of the neighbor space segment, the load of node A cannot be balanced
[0007] It can be seen that due to the problem that node load distribution is affected by neighbor nodes, the current distributed keyword table network either cannot achieve load balancing; or needs to rely on very complex algorithms and sacrifice performance to frequently adjust the segmentation and allocation of keyword space to achieve Load balancing; or it is necessary to divide the keyword space into very many segments at the cost of greatly increasing the routing table, and allocate many segments of keyword space to each node, resulting in cumbersome maintenance and use of the routing table. In short, the existing distributed The keyword table network cannot effectively achieve load balancing

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A Distributed Keyword Table Network and Its Operation Method
  • A Distributed Keyword Table Network and Its Operation Method
  • A Distributed Keyword Table Network and Its Operation Method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0046] figure 2 It is a schematic diagram of keyword space segmentation of a keyword table network in Embodiment 1 of the present invention, such as figure 1 As shown, a distributed keyword table network of the present invention includes multiple identical keyword spaces. Data / tasks are stored / allocated at most one copy per keyspace. A key space can be divided into multiple segments, and the space is assigned to multiple nodes in segments. A key space can also be undivided and completely assigned to a node.

Embodiment 2

[0048] image 3 It is a schematic diagram of a keyword table network load distribution in Embodiment 2 of the present invention, such as image 3 As shown, the keyword table network has two identical keyword spaces, and the two keyword spaces have three nodes, respectively A1, B1, C1, and A2, B2, C2, nodes A1, B1, C1, And A2, B2, and C2 are allocated a keyword space segment in the two spaces respectively. How long each node is responsible for in the two spaces of the key space segment is not affected by other nodes. Suppose the capacity ratio of nodes A1, B1, and C1 is 3:6:7, and the capacity ratio of nodes A2, B2, and C2 is also 3:6:7. To achieve load balancing, nodes A1, B1, C1, and A2, The length ratio of the space segments B2 and C2 are responsible for in each space is also 3:6:7. In this case, no matter whether the two spaces are equal spaces, or a primary space and a backup space, the loads of the three nodes are balanced of.

Embodiment 3

[0050] Figure 4 It is a schematic diagram of a keyword table network load distribution in Embodiment 3 of the present invention, such as Figure 4 As shown, the keyword table network has two identical keyword spaces, and there are four nodes A, B, C, and D in total. Nodes A and B are allocated a keyword space segment in space 1 respectively, and nodes C and D are respectively A key space segment is allocated in space 2. Each node is responsible for how long the key space segment is not affected by other nodes. The nodes in the two spaces are completely different to facilitate highly reliable multi-region or multi-data center data storage. The nodes of space 1 and space 2 are different and distributed in different regions or data centers, so that each data can be stored in two regions or data centers, avoiding the loss of both data due to regional failures.

[0051] refer to Figure 4 , assuming that the two spaces are equal, and there is no distinction between master and ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a distributed keyword table network. The distributed keyword table network maintains a plurality of keyword spaces with the same keyword value range. Each keyword space is divided into at least one keyword space segment. The keyword space Segments are assigned to nodes, and each node is responsible for at least one keyspace segment in at least one keyspace. The invention also correspondingly discloses a method for operating the distributed keyword table network. Through the present invention, each node in the distributed keyword table network can flexibly allocate data / tasks, and the nodes do not affect each other in task / data allocation, so that load allocation is flexible, and load balancing can be effectively realized at a relatively small cost.

Description

technical field [0001] The invention relates to the field of computer networks, in particular to a distributed keyword table network and a method for operating it. Background technique [0002] Distributed keyword (key) table is a general term for two very similar P2P technologies, distributed hash table (distributed hash table, DHT) and distributed ordering table (distributed ordering table). Distributed hash tables and distributed sequence tables have become hot technologies in the P2P field in recent years, and are applied to various P2P applications such as BT and eMule. [0003] The overlay network constructed with distributed keyword table technology is called a distributed keyword table network, including a distributed hash table network and a distributed sequence table network. The distributed keyword table network can provide the function of data distributed storage and management, and can also provide the function of distributed task processing and management. Cl...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Patents(China)
IPC IPC(8): H04L29/08G06F17/30
Inventor 李漓春
Owner ZTE CORP