A system and method for processing multi-stage distributed task scheduling

A technology of distributed task and scheduling method, which is applied in the system field of multi-stage distributed task scheduling, and can solve problems such as not reflecting the idea of ​​phased processing of complex transactions, high overhead, and inability to implement sub-stage rollback and redo, etc. Achieve efficient and reliable processing, no aftereffect, and reduce costs

Active Publication Date: 2017-03-22
SHANGHAI STOCK EXCHANGE
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The overall processing method ensures the integrity of the transaction, but once a rollback occurs, the overall processing overhead will be very high, and the existing distributed two-stage transaction processing divides the distributed transaction into two stages: The preparation phase of the first phase and the commit phase of the second phase. In the preparation phase, each sub-transaction can be processed in a distributed manner, and each sub-transaction has completed almost all of the formal submission work. In the commit phase, the entire transaction is submitted or returned. Roll operation, distributed two-stage transaction processing also ensures the integrity of the transaction, but this mechanism also requires all subtasks to be done or not to be done, and rollback and redo at the sub-stage level cannot be realized. At the same time, the two-stage The transaction processing method does not reflect the staged processing idea of ​​complex transactions

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 system and method for processing multi-stage distributed task scheduling
  • A system and method for processing multi-stage distributed task scheduling
  • A system and method for processing multi-stage distributed task scheduling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] Such as figure 1 as shown, figure 1 It is a schematic diagram of the system deployment method for processing multi-stage distributed task scheduling. In the figure, a cluster composed of two hosts is taken as an example. Synchronization locks are used between cluster nodes to support functions such as master node election and inter-node control message triggering. All The transactions sent from the front desk are routed to the architecture module of the master node through the communication layer. The architecture module of the master node records the transaction data into the working page log file of the shared disk, and triggers the replay of the working page log file by the slave node architecture module through a synchronization lock , the architecture module of the master-slave node will send its own transaction request to the function module of the local machine for processing, and the response information processed by the function module will be summarized to the...

Embodiment 2

[0035] figure 2 It is a flowchart of the cooperative work of the master-slave node architecture module and its functional modules during the multi-stage processing of transaction requests. As shown in the figure, all transaction requests are first processed by the master node architecture module, and the master node architecture module completes the necessary calibration. After verifying the work, create an asynchronous job for the transaction request and send the asynchronous transaction job to the function module for execution. After that, the master node architecture module immediately triggers the slave node architecture module to replay the work page log file, and the slave node architecture module is based on the log The data recorded in the file creates the same transaction asynchronous job and request data locally, and sends the job request to the local function module for processing. The function module of each node receives the transaction request through an asynchro...

Embodiment 3

[0039] image 3 This is the flow chart of the takeover process of the standby machine for node failure in the cluster. If a node fails and exits in the cluster, the synchronization message lock between the cluster nodes will trigger a node exit event on all nodes, and the remaining nodes in the cluster will respond to the event. If the master node fails, the newly elected master node will replay according to the working page log file recorded by the failed master node on the shared disk, restore to the same state as the last record of the failed master node, and then continue to take over the task processing; if If the slave node fails, the standby machine will directly take over the work of the failed slave node, and the master node will resend unfinished transactions to the slave node for processing. This mechanism ensures that multi-stage distributed complex transactions can still be processed correctly even when the master node fails. .

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 the data processing field and specifically relates to a system and a method used for processing multi-phase distributed task scheduling. The system comprises multiple host computers for specific business processing, each host computer is for set processing of one or more products, sharing access to a disk is realized through a storing area network among the host computers, data synchronization among the host computers is carried out through a sharing disc, message synchronization among all the transaction host computers is realized through distributed locks, a function module is for classification of request processing of a present machine product set, distributed subtransaction processing is mainly accomplished at the function module of each host computer. Compared with the prior art, the system is advantaged in that processing efficiency of a multi-phase distributed complex transaction is improved through a distribution mode, rollback cost of the transaction caused by processing failure is reduced, and the system is especially suitable for high-efficiency reliable processing of the multi-stage, no-aftereffect and sub-phase-distribution complex transaction in the security data processing industry.

Description

[0001] [technical field] [0002] The present invention relates to the field of data processing, in particular to a processing method suitable for multi-stage distributed complex transaction processing requirements of a high-availability cluster computer system, in particular to a system and method for processing multi-stage distributed task scheduling. [0003] [Background technique] [0004] At present, in the field of securities data processing, we are faced with a type of complex business called multi-stage distributed complex transactions, such as ETF (Exchange Tradable Fund, Exchange Traded Fund) purchase and redemption, simultaneous trading of a basket of stocks and other data processing requirements. This type of business often has the following characteristics: In the time dimension, it can be divided into multiple natural sub-phase transactions that are relatively independent in time sequence. If and only when the current sub-phase transaction is completed, the next su...

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): H04L29/08
Inventor 龚伟武剑锋王泊楼晓鸿林志高黄寅飞吴征徐乾白硕
Owner SHANGHAI STOCK EXCHANGE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap