Software defined network publish-subscribe system and method

Patent metadata

Field	Value
Publication number	US-11729133-B2
Application number	US-202217966759-A
Country	US
Kind code	B2
Filing date	Oct 14, 2022
Priority date	Apr 20, 2020
Publication date	Aug 15, 2023
Grant date	Aug 15, 2023

Abstract

Embodiments of the present disclosure provide a software defined network publish-subscribe system and method. The software defined network publish-subscribe system comprises a controller, an administrator controller, a sequencing subsystem and a message multi-backup cache subsystem. The controller is configured to process publish-subscribe transactions in clusters, and comprises sequencing nodes used for generating ordered serial numbers for messages and cache nodes for caching the messages. The administrator controller is configured to simultaneously perform functions of the controller and manage software-defined network. The sequencing subsystem is configured to realize the cooperative sequencing between the administrator controller and the sequencing nodes. The message multi-backup cache subsystem is configured to realize the consistency of the messages through the administrator controller and the cache nodes. Through the interaction of various modules, the consensus of various subscribers on the messages and the message serial numbers is realized on the software defined network publish-subscribe system.

First claim

What is claimed is:

1. A software defined network publish-subscribe system, comprising:
a processor;
a controller, an administrator controller, a sequencing subsystem and a message multi-backup cache subsystem, wherein
the controller is configured to process publish-subscribe transactions in clusters, and the controller comprises sequencing nodes used for generating ordered serial numbers for messages and cache nodes for caching the messages;
the administrator controller is configured to simultaneously perform functions of the controller and managing software-defined network;
the sequencing subsystem is configured to realize a cooperative sequencing between the administrator controller and the sequencing nodes;
the message multi-backup cache subsystem is configured to realize consistency of the messages through the administrator controller and the cache nodes;
wherein the message multi-backup cache subsystem comprises:
a cache node selection module configured to select the cache nodes comprising temporary cache nodes, network edge cache nodes and backup cache nodes;
a message cache management module configured to realize a persistent storage of the messages, and adopt a double-layer index mode to realize a search of the messages;
a node-to-node message synchronization module configured to synchronize the messages among the temporary cache nodes, the network edge cache nodes and the backup cache nodes;
the cache node selection module is further configured to:
define publishing nodes of the messages as the temporary cache nodes;
divide publishing clusters into no less than one publishing group according to a publish-subscribe situation and a physical connection situation of the system, and select one central publishing cluster in each publishing group as the network edge cache node;
select no less than one cluster node on a message forwarding path of each publishing group as a backup cache node of the network edge cache node;
the node-to-node message synchronization module is further configured to:
send the messages in batches to the network edge cache nodes through the temporary cache nodes;
match ID numbers of the messages with serial numbers through the network edge cache nodes, cache the messages in files according to a serial number sequence after successful matching, and send a paired ID number and a serial number list to the backup cache nodes;
calculate current missing messages through the backup cache nodes, and return the serial numbers of the missing messages thereof to the network edge cache nodes; and
synchronize the missing messages of the backup cache nodes through the network edge cache nodes.

2. The system according to claim 1 , wherein the sequencing subsystem is configured to:
drive the sequencing nodes in the controller to sequence the messages according to administrator tokens, and sequence the messages in batches in a distributed sequencing mode when the sequencing nodes are multi-sequencing nodes; and
the sequencing nodes issue the serial numbers in a secondary broadcast mode.

3. The system according to claim 2 , wherein the sequencing subsystem specifically comprises:
a sequencing node selection module configured to realize a selection of the sequencing nodes through the administrator controller;
a sequencing node crash management module configured to realize a crash management of the sequencing nodes through the administrator controller;
an administrator controller crash management module configured to realize a crash management of the administrator controller through the sequencing nodes; and
an administrator controller reselection module configured to realize a reselection of the administrator controller through the sequencing nodes.

4. The system according to claim 3 , wherein the sequencing node selection module is configured to:
define one sequencing node for each subject tree by taking the subject tree as a unit;
monitor the messages according to the sequencing nodes when the publishing nodes publish the messages;
the sequencing node crash management module is further configured to: select the sequencing nodes through a sequencing node selection algorithm according to a topology situation of each cluster in a network and a publishing situation of each cluster node, and
send the selected sequencing nodes to the administrator controller through broadcasting;
wherein the administrator controller determines whether the sequencing nodes crash or not according to replies received from the sequencing nodes.

5. The system according to claim 3 , wherein the administrator controller crash management module is further configured to:
start a sequencing thread, and send a successful starting message to the administrator controller, and
determine whether the administrator controller crashes or not according to replies received from the administrator controller;
the administrator controller reselection module is further configured to:
send a voting application to other sequencing nodes except a present sequencing node, and vote for the present sequencing node; and
select a new administrator controller according to received messages sent by other sequencing nodes.

6. A software defined network publish-subscribe method, comprising the following steps:
sending administrator tokens to sequencing nodes;
sequencing messages by the sequencing nodes, wherein the messages are sequenced in batches in a distributed sequencing mode when the sequencing nodes are multi-sequencing nodes;
selecting the sequencing nodes according to a topology situation of each cluster in a network and a publishing situation of each cluster node, and sending the selected sequencing nodes identifiers through broadcasting;
determining whether the sequencing nodes have crashed or not according to replies received from the sequencing nodes;
sending a message of success in starting a thread, and determining whether an administrator controller crashes or not according to received feedback information;
defining publishing nodes of the messages as temporary cache nodes, selecting central publishing clusters as network edge cache nodes from publishing groups, and selecting cluster nodes on message forwarding paths of the publishing groups as backup cache nodes of the network edge cache nodes;
persistently storing the messages through the network edge cache nodes;
sending the messages in batches to the network edge cache nodes through the temporary cache nodes;
matching ID numbers of the messages with serial numbers through the network edge cache nodes, and caching the messages in files according to a serial number sequence; and
calculating current missing messages through the backup cache nodes, and synchronizing the missing messages of the backup cache nodes to the backup cache nodes through the network edge cache nodes.

Assignees

• Univ Beijing Posts & Telecomm

Inventors

• Cheng Bo
• Zhang Yang
• Zhao Shuai
• Zhang Yingying
• Chen Junliang

Classifications

• H04L51/58 Primary

  Message adaptation for wireless communication · CPC title

• H04L67/568

  Storing data temporarily at an intermediate stage, e.g. caching · CPC title

• H04L67/10 Primary

  in which an application is distributed across nodes in the network (software deployment G06F8/60; multiprogramming arrangements G06F9/46) · CPC title

• H04L67/55 Primary

  Push-based network services · CPC title

• H04L67/5682

  Policies or rules for updating, deleting or replacing the stored data · CPC title