With the transformation of the financial services market, the demand for core technology for conducting financial transactions has grown. The financial transaction model has changed from traditional manual bidding to matchmaking orders through an electronic transaction system. This article is a follow up article to the initial article which detailed the matchmaking technology architectural design, and aims to answer frequently asked questions on the topic.
Q: On a hot standby mechanism. How can one ensure strong consistency of status on multiple machine memories?
A: All hot standby machines are stateless. The system deploys a group of matchmaking engines into a mutual standby matchmaking engine cluster. Additionally, there is only one matchmaking engine in service at any given time. When one of the operating matchmaking engines breaks down and cannot continue to work, the matchmaking engine cluster will quickly detect such faults and select a standby matchmaking engine to take over the task of the faulty matchmaking engine. Multi-machine hot standby technology behaves like a standby technology to resolve the temporary fault of the server.
Q: If one adds a new matchmaking engine, how can we judge that the synchronization of all matchmaking data into the new engine has occurred?
A: Ideally, the engine writes the raid document system into the commission sheet so that one can retrieve the data in case of a new matchmaking engine.
Q: According to your description, is it true that a cluster shall at least be fitted with three matchmaking engines to judge if the matchmaking engine breaks down?
A: The matchmaking progress shall be monitored by risk control process.
Q: Which one do the memory state machine replicate, the results or the data?
A: The data. It means to replicate the data with a certain status.
Q: If there are only two matchmaking engines and we discover that the results are inconsistent with a buying order. How can we judge which is the faulty engine?
A: Only one machine conducts matchmaking. It writes back to the database, produces a quotation and distributes to each terminal only after the completion of matchmaking. After that the user can check if the system has executed their commission sheet.
2,605 posts | 747 followers
FollowSamuel_Song - December 9, 2020
Alibaba Clouder - March 8, 2017
Alibaba Clouder - September 1, 2020
Alibaba Clouder - February 14, 2020
Alibaba Cloud Community - February 28, 2022
Alibaba Cloud Native Community - November 15, 2023
2,605 posts | 747 followers
Follow
Lindorm
Lindorm is an elastic cloud-native database service that supports multiple data models. It is capable of processing various types of data and is compatible with multiple database engine, such as Apache HBase®, Apache Cassandra®, and OpenTSDB.
Learn MoreMore Posts by Alibaba Clouder
