CN114283001A - Matching system based on memory algorithm - Google Patents
Matching system based on memory algorithm Download PDFInfo
- Publication number
- CN114283001A CN114283001A CN202111622475.0A CN202111622475A CN114283001A CN 114283001 A CN114283001 A CN 114283001A CN 202111622475 A CN202111622475 A CN 202111622475A CN 114283001 A CN114283001 A CN 114283001A
- Authority
- CN
- China
- Prior art keywords
- matching
- queue
- order
- price
- selling
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Abstract
The invention provides a matching system based on a memory algorithm, which comprises: a transaction layer: the client terminal comprises a mobile APP, a PC end and a Web end, and is provided with a programming interface, so that a final user can perform operations of matching entrustment, entrustment inquiry and deposit and withdrawal through the client terminal; interface layer: the system comprises a gateway cluster, a business layer and a transaction layer, wherein the gateway cluster is provided with a product rule forwarding module, and requests of users for commodity inquiry, order placement and the like are distributed to the business layer and the generated quotation is fed back to the transaction layer; and (4) a service layer: the system comprises a plurality of matching engine clusters which are deployed by matching engines and are used for receiving orders, realizing order matching according to business logic and generating transaction records at the same time, and then giving feedback to users about transaction results; and (3) a data layer: the system comprises a database and a file system, and is provided with an asynchronous data persistence module for storing commodity information, transaction information, fund information and user information and performing persistence. The system has the characteristics of high performance, high reliability and high expansibility.
Description
Technical Field
The invention relates to the technical field of electronic transaction systems, in particular to a matching system based on a memory algorithm.
Background
The traditional financial transaction mainly occurs in a certain financial market, buyers and sellers of the financial transaction carry out price negotiation in modes of price calling and the like, a transaction is formed after final agreement is achieved, and meanwhile, real object delivery is carried out to a specified delivery place according to a transaction order. With the increasing information technology and the rapid development of financial services, the market demand for financial transactions is increasing, and the mainstream financial transaction mode has been changed from the traditional manual price calling mode to the mode of matching orders by a highly electronic transaction system. Since the whole process of transaction is mainly performed by people, the conventional financial transaction has the following disadvantages: the method has the advantages of low efficiency, low speed, large transaction time limit, large transaction space limit, high transaction cost, easy on-screen transaction, poor transaction expansibility, easy error of transaction, poor fund safety and the like.
Disclosure of Invention
In order to overcome the defects of the prior art, the invention provides a matching system based on a memory algorithm, which has the characteristics of high performance, high reliability and high expansibility.
In order to achieve the above object, the present invention provides a matching system based on a memory algorithm, comprising:
a transaction layer: the client terminal comprises a mobile APP, a PC end and a Web end, and is provided with a programming interface, so that a final user can perform operations of matching entrustment, entrustment inquiry and deposit and withdrawal through the client terminal;
interface layer: the system comprises a gateway cluster, a business layer and a transaction layer, wherein the gateway cluster is provided with a product rule forwarding module, and requests of users for commodity inquiry, order placement and the like are distributed to the business layer and the generated quotation is fed back to the transaction layer;
and (4) a service layer: the system comprises a plurality of matching engine clusters which are deployed by matching engines and are used for receiving orders, realizing order matching according to business logic and generating transaction records at the same time, and then giving feedback to users about transaction results;
and (3) a data layer: the system comprises a Database (DB) and a file system, and is provided with an asynchronous data persistence module for storing commodity information, transaction information, fund information and user information and carrying out persistence, and a cache stores quotations to reduce the pressure of the database.
Furthermore, the matching engine comprises an atomic multicast order sequencing module, a matching processing module, a transaction record log module and memory data, and a matching logic process of the matching engine runs on a memory of the server;
the atomic multicast order sequencing module comprises a trade order receiving thread, a trade order sending thread and a trade information sending thread;
the matching processing module comprises a peripheral service logic thread, a matching service logic thread, a transaction quotation issuing thread and a transaction quotation issuing thread;
the transaction record log module comprises a synchronous log write thread and an asynchronous persistent agent thread;
the memory data comprises order information and transaction information.
Further, the matching engine matches the buy and sell orders to form a transaction record finally, the matching engine establishes a buy queue and a sell queue for the orders which cannot be matched immediately so as to store the orders, the buy queue and the sell queue are sorted according to the price priority and the time priority principle of the same price, the buy queue is sorted according to the order of the entrusted price from high to low, and the sell queue is sorted according to the order of the entrusted price from low to high.
Further, the matching engine carries out classification processing on the new transaction entrusts of the trade orders, and the processing mode comprises the following steps:
when new is to buy a market price commission: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase entrusts have the tradable quantity after all the market price entrusts in the selling queue are matched, the market price entrusts are matched with the limit price entrusts in the selling queue; if no market price entrusts exist in the selling queue, directly matching with the price limit entrusts in the selling queue; if the sell queue is empty, simply insert the buy order into the buy queue;
when the new purchase limit order: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase order has the tradable quantity and the price is more than or equal to the lowest price of the selling in the selling queue after the market price order is matched completely, the purchase order is matched with the price limit order continuously; if only a price limit entrust exists in the selling queue and the price of the buying entrust is more than or equal to the lowest price of selling one in the selling queue, matching the entrusts, and otherwise, only inserting the buying entrust into the buying queue according to a matching principle; if the sell queue is empty, the buy order is directly inserted into the buy queue according to the matching principle;
when newly entered is to sell the market price entrustment: if the buying queue has a market price commission, matching is carried out; if the sale queue has outstanding volume after matching, matching with the limit order in the buy queue; if only the limited price entrusts in the buying queue, matching the limited price entrusts; inserting the sell order into a sell queue if the buy queue is empty;
when the new order is a selling limit order: if the purchase queue has market price entrusts, matching the purchase queue with the market price entrusts firstly; if the market price entrusts have the tradable quantity and the price is less than or equal to the highest price bought in the buy queue after all the market price entrusts are matched, the market price entrusts continue to be matched with the price limit entrusts; if only a limited price entrustment exists in the buying queue and the price of the selling entrustment is less than or equal to the highest price bought in the buying queue, matching is carried out, otherwise, the selling entrustment can only be inserted into the selling queue according to a matching principle; if the buy queue is empty, the sell order is inserted directly into the sell queue on a match basis.
Further, the matching order of the matching engine is as follows:
when the matching engine receives a new purchase order, searching whether a purchase order meeting the price rule exists at the head of the selling queue, and if the purchase order with the selling price less than or equal to the purchase price exists, taking the order out of the selling queue and matching into a transaction; if the selling queue is empty or the head of the selling queue does not meet the price relationship, inserting the buy order into the buy queue, and inserting the newly inserted buy order into the buy queue after one-time sequencing;
when a matching engine receives a new selling order, checking whether a buying order meeting a price rule exists at the head of a buying queue, and if the buying order with the buying price greater than or equal to the selling price exists, taking the order out of the order queue and matching the order into a transaction; and if the buy queue is empty or the head of the buy queue does not meet the price relationship, inserting the sell order into the sell queue, and inserting the newly inserted sell order into the sell queue after one-time sequencing.
Furthermore, the matching engine deploys a plurality of hot backups of the matching engine cluster in a mutual backup mode, one matching engine provides service in the same time, and when the matching engine fails and cannot continue to work normally, the matching engine cluster can quickly detect the failure and select a backup matching engine to take over the task of the failed matching engine, so that the normal operation of the whole matching system is ensured.
Furthermore, the running matching algorithm is defined as a deterministic state machine by the matching engine cluster in a memory state machine copying mode and is copied to be deployed in a plurality of matching engines of the matching engine cluster, matching algorithm copies on each matching engine start to run from the same initial state, when the matching engine cluster receives a purchase order sent by the gateway cluster, each matching engine can match the purchase order and sequentially generate transaction records, meanwhile, the independent state of the deterministic matching algorithm state machine is updated, when the matching engine cluster runs normally, the matching algorithm copies on each matching engine have the same result state, and when the matching engine cluster fails or is abnormal, the matching engines can generate state inconsistency, so that the system abnormality is judged.
Further, the mode of copying the memory state machine comprises the steps of solving reliable multicast and global ordering of the matching engine orders by adopting atomic multicast, providing quick order matching by adopting a pipelining matching technology based on a lock-free order queue, realizing interaction with a database by adopting an asynchronous consistency persistence technology, monitoring the state of the matching engine cluster by adopting a failure backup technology and ensuring the fault-tolerant capability of the system, and solving the recovery of a fault matching engine or the addition of a new matching engine by adopting a progress pursuit technology.
Furthermore, the database establishes an index in the order table to improve the query efficiency, and the index is a single index.
The matching system based on the memory algorithm is based on a multilayer distributed system, is designed and realized by using a J2EE technology, adopts the latest memory matching technology, improves the system operation efficiency by using a multilevel storage mode, and constructs a transaction matching system with high performance, high reliability and high expansibility.
Drawings
The present invention will be further described and illustrated with reference to the following drawings.
FIG. 1 is a system block diagram of a memory algorithm based matching system in accordance with a preferred embodiment of the present invention;
FIG. 2 is a software architecture diagram of a memory algorithm based matching system;
FIG. 3 is a hardware framework diagram of a matching system based on memory algorithm according to the preferred embodiment of the present invention
FIG. 4 is a schematic diagram of the ordering principles of a buy queue and a sell queue;
FIG. 5 is a flow diagram of a match engine making a match;
FIG. 6 is a schematic diagram of a matchmaking engine cluster multi-machine hot backup format.
Detailed Description
The technical solution of the present invention will be more clearly and completely explained by the description of the preferred embodiments of the present invention with reference to the accompanying drawings.
As shown in fig. 1 and fig. 2, a matching system based on a memory algorithm according to a preferred embodiment of the present invention includes:
a transaction layer: the client terminal comprises a mobile APP, a PC end and a Web end, and is provided with a programming interface, so that a final user can perform operations of matching entrustment, entrustment inquiry and deposit and withdrawal through the client terminal;
interface layer: the system comprises a gateway cluster, a business layer and a transaction layer, wherein the gateway cluster is provided with a product rule forwarding module, and requests of users for commodity inquiry, order placement and the like are distributed to the business layer and the generated quotation is fed back to the transaction layer;
and (4) a service layer: the system comprises a plurality of matching engine clusters which are deployed by matching engines and are used for receiving orders, realizing order matching according to business logic and generating transaction records at the same time, and then giving feedback to users about transaction results;
and (3) a data layer: the system comprises a Database (DB) and a file system, and is provided with an asynchronous data persistence module, the database establishes indexes in an order form to improve the query efficiency, the indexes are single indexes and are used for storing commodity information, transaction information, fund information and user information and carrying out persistence, and the cache stores quotations to reduce the pressure of the database.
The matching module is divided into a plurality of service partitions according to different types of financial transaction exhibits, and each partition is independently matched without influencing each other.
Specifically, the method comprises the following steps: as shown in fig. 3, a hardware architecture of a matching system based on a memory algorithm of the present invention may be composed of N clients, N gateways, X product clusters (each cluster is composed of 2 to 3 matching engines and is responsible for processing in response to a product order), a transaction record database, and an optional monitoring system. The client is connected to the corresponding gateway, and the gateway is responsible for receiving the order submitted by the client and forwarding the order to the corresponding product cluster according to the financial product category related to the order. All matching engines in the product cluster receive orders sent by the gateway, match the orders and feed back messages to the gateway and the client according to matching business rules, and meanwhile, transaction records generated by matching are persisted in a transaction record database.
The matching engine comprises an atomic multicast order sequencing module, a matching processing module, a transaction record log module and memory data, and a matching logic process of the matching engine runs on a memory of the server;
the atomic multicast order sequencing module comprises a trade order receiving thread, a trade order sending thread and a trade information sending thread;
the matching processing module comprises a peripheral service logic thread, a matching service logic thread, a transaction quotation issuing thread and a transaction quotation issuing thread;
the transaction record log module comprises a synchronous log write thread and an asynchronous persistent agent thread;
the memory data comprises order information and transaction information.
As shown in fig. 4, the matching engine matches the buy and sell orders to form a transaction record, the matching engine establishes a buy queue and a sell queue for storing orders that cannot be matched immediately, the buy queue and the sell queue are sorted according to the price-first and price-under-time-first principles, the buy queue is sorted according to the order of the entrusted price from high to low, and the sell queue is sorted according to the order of the entrusted price from low to high.
The matching engine carries out classification processing on the new transaction entrusts of the trade orders, and the processing mode comprises the following steps:
when new is to buy a market price commission: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase entrusts have the tradable quantity after all the market price entrusts in the selling queue are matched, the market price entrusts are matched with the limit price entrusts in the selling queue; if no market price entrusts exist in the selling queue, directly matching with the price limit entrusts in the selling queue; if the sell queue is empty, simply insert the buy order into the buy queue;
when the new purchase limit order: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase order has the tradable quantity and the price is more than or equal to the lowest price of the selling in the selling queue after the market price order is matched completely, the purchase order is matched with the price limit order continuously; if only a price limit entrust exists in the selling queue and the price of the buying entrust is more than or equal to the lowest price of selling one in the selling queue, matching the entrusts, and otherwise, only inserting the buying entrust into the buying queue according to a matching principle; if the sell queue is empty, the buy order is directly inserted into the buy queue according to the matching principle;
when newly entered is to sell the market price entrustment: if the buying queue has a market price commission, matching is carried out; if the sale queue has outstanding volume after matching, matching with the limit order in the buy queue; if only the limited price entrusts in the buying queue, matching the limited price entrusts; inserting the sell order into a sell queue if the buy queue is empty;
when the new order is a selling limit order: if the purchase queue has market price entrusts, matching the purchase queue with the market price entrusts firstly; if the market price entrusts have the tradable quantity and the price is less than or equal to the highest price bought in the buy queue after all the market price entrusts are matched, the market price entrusts continue to be matched with the price limit entrusts; if only a limited price entrustment exists in the buying queue and the price of the selling entrustment is less than or equal to the highest price bought in the buying queue, matching is carried out, otherwise, the selling entrustment can only be inserted into the selling queue according to a matching principle; if the buy queue is empty, the sell order is inserted directly into the sell queue on a match basis.
As shown in fig. 5, the matching order of the matching engine is:
when the matching engine receives a new purchase order, searching whether a purchase order meeting the price rule exists at the head of the selling queue, and if the purchase order with the selling price less than or equal to the purchase price exists, taking the order out of the selling queue and matching into a transaction; if the selling queue is empty or the head of the selling queue does not meet the price relationship, inserting the buy order into the buy queue, and inserting the newly inserted buy order into the buy queue after one-time sequencing;
when a matching engine receives a new selling order, checking whether a buying order meeting a price rule exists at the head of a buying queue, and if the buying order with the buying price greater than or equal to the selling price exists, taking the order out of the order queue and matching the order into a transaction; and if the buy queue is empty or the head of the buy queue does not meet the price relationship, inserting the sell order into the sell queue, and inserting the newly inserted sell order into the sell queue after one-time sequencing.
In combination with the trade order situation, the orderliness of the trade queue is the basis for ensuring the certainty of the matching sequence, and each order can match the current optimal transaction in the matching process.
As shown in fig. 6, the matching engine deploys a plurality of hot backups of the matching engine cluster in a mutual backup manner, and one matching engine provides service at the same time, and when the matching engine fails and cannot continue to work normally, the matching engine cluster can quickly detect the failure and elect a backup matching engine to take over the task of the failed matching engine, thereby ensuring the normal operation of the whole matching system. Through a multi-machine hot backup technology, the switching between the host machine and the standby machine is completed within an acceptable time, and the standby machine provides seamless continuous service, so that the long-time matching service interruption is avoided, and the long-time and reliable service of a matching system is ensured. In an actual production environment, three servers which are backups of each other can provide higher reliability which can be used for the production environment.
The running matching algorithm is defined as a deterministic state machine by a matching engine cluster in a memory state machine copying mode and is copied to be deployed into a plurality of matching engines of a matching engine cluster, matching algorithm copies on each matching engine start to run from the same initial state, when the matching engine cluster receives a trading order sent by a gateway cluster, each matching engine can match the trading order and sequentially generate a transaction record, the independent state of the deterministic matching algorithm state machine is updated at the same time, when the matching engine cluster normally runs, the matching algorithm copies on each matching engine have the same result state, and when the matching engine cluster fails or is abnormal, the matching engines can generate state inconsistency so as to judge that the system is abnormal.
The mode of copying the memory state machine comprises the steps of solving reliable multicast and global orderliness of matching engine orders by adopting atomic multicast, providing rapid order matching by adopting a production line matching technology based on a lock-free order queue, realizing interaction with a database by adopting an asynchronous consistency persistence technology, carrying out state monitoring on a matching engine cluster by adopting a fail-over technology and ensuring the fault-tolerant capability of a system, and solving the recovery of a fault matching engine or the addition of a new matching engine by adopting a progress catch-up technology.
The matching system based on the memory algorithm is based on a multilayer distributed system, is designed and realized by using a J2EE technology, adopts the latest memory matching technology, improves the system operation efficiency by using a multilevel storage mode, and constructs a transaction matching system with high performance, high reliability and high expansibility.
The above detailed description merely describes preferred embodiments of the present invention and does not limit the scope of the invention. Without departing from the spirit and scope of the present invention, it should be understood that various changes, substitutions and alterations can be made herein by those skilled in the art without departing from the spirit and scope of the invention as defined by the appended claims and their equivalents. The scope of the invention is defined by the claims.
Claims (9)
1. A matching system based on a memory algorithm, comprising:
a transaction layer: the client terminal comprises a mobile APP, a PC end and a Web end, and is provided with a programming interface, so that a final user can perform operations of matching entrustment, entrustment inquiry and deposit and withdrawal through the client terminal;
interface layer: the system comprises a gateway cluster, a business layer and a transaction layer, wherein the gateway cluster is provided with a product rule forwarding module, and requests of users for commodity inquiry, order placement and the like are distributed to the business layer and the generated quotation is fed back to the transaction layer;
and (4) a service layer: the system comprises a plurality of matching engine clusters which are deployed by matching engines and are used for receiving orders, realizing order matching according to business logic and generating transaction records at the same time, and then giving feedback to users about transaction results;
and (3) a data layer: the system comprises a Database (DB) and a file system, and is provided with an asynchronous data persistence module for storing commodity information, transaction information, fund information and user information and carrying out persistence, and a cache stores quotations to reduce the pressure of the database.
2. The memory algorithm-based matching system of claim 1, wherein the matching engine comprises an atomic multicast order sequencing module, a matching processing module, a transaction record log module and memory data, and a matching logic process of the matching engine runs on a memory of the server;
the atomic multicast order sequencing module comprises a trade order receiving thread, a trade order sending thread and a trade information sending thread;
the matching processing module comprises a peripheral service logic thread, a matching service logic thread, a transaction quotation issuing thread and a transaction quotation issuing thread;
the transaction record log module comprises a synchronous log write thread and an asynchronous persistent agent thread;
the memory data comprises order information and transaction market information.
3. The memory algorithm based matching system of claim 1, wherein the matching engine matches buy and sell orders to form a transaction record, the matching engine establishes a buy queue and a sell queue for orders that cannot be matched immediately to store the orders, the buy queue and the sell queue are sorted according to a price-first and a price-agreement time-first principle, the buy queue is sorted according to an order of an order price from high to low, and the sell queue is sorted according to an order of an order price from low to high.
4. The memory algorithm based matching system of claim 3, wherein the matching engine classifies incoming transaction orders for the trade orders in a manner that includes:
when new is to buy a market price commission: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase entrusts have the tradable quantity after all the market price entrusts in the selling queue are matched, the market price entrusts are matched with the limit price entrusts in the selling queue; if no market price entrusts exist in the selling queue, directly matching with the price limit entrusts in the selling queue; if the sell queue is empty, simply insert the buy order into the buy queue;
when the new purchase limit order: if the market price entrusts in the selling queue, matching the market price entrusts; if the purchase order has the tradable quantity and the price is more than or equal to the lowest price of the selling in the selling queue after the market price order is matched completely, the purchase order is matched with the price limit order continuously; if only a price limit entrust exists in the selling queue and the price of the buying entrust is more than or equal to the lowest price of selling one in the selling queue, matching the entrusts, and otherwise, only inserting the buying entrust into the buying queue according to a matching principle; if the sell queue is empty, the buy order is directly inserted into the buy queue according to the matching principle;
when newly entered is to sell the market price entrustment: if the buying queue has a market price commission, matching is carried out; if the sale queue has outstanding volume after matching, matching with the limit order in the buy queue; if only the limited price entrusts in the buying queue, matching the limited price entrusts; inserting the sell order into a sell queue if the buy queue is empty;
when the new order is a selling limit order: if the purchase queue has market price entrusts, matching the purchase queue with the market price entrusts firstly; if the market price entrusts have the tradable quantity and the price is less than or equal to the highest price bought in the buy queue after all the market price entrusts are matched, the market price entrusts continue to be matched with the price limit entrusts; if only a limited price entrustment exists in the buying queue and the price of the selling entrustment is less than or equal to the highest price bought in the buying queue, matching is carried out, otherwise, the selling entrustment can only be inserted into the selling queue according to a matching principle; if the buy queue is empty, the sell order is inserted directly into the sell queue on a match basis.
5. The memory algorithm-based matching system of claim 3, wherein the matching order of the matching engine is as follows:
when the matching engine receives a new purchase order, searching whether a purchase order meeting the price rule exists at the head of the selling queue, and if the purchase order with the selling price less than or equal to the purchase price exists, taking the order out of the selling queue and matching into a transaction; if the selling queue is empty or the head of the selling queue does not meet the price relationship, inserting the buy order into the buy queue, and inserting the newly inserted buy order into the buy queue after one-time sequencing;
when the matching engine receives a new selling order, checking whether a buying order meeting the price rule exists at the head of the buying queue, and if the buying order with the buying price larger than or equal to the selling price exists, taking the order out of the order queue and matching the order into a transaction; and if the buy queue is empty or the head of the buy queue does not meet the price relationship, inserting the sell order into the sell queue, and inserting the newly inserted sell order into the sell queue after one-time sequencing.
6. The memory algorithm-based matching system of claim 1, wherein the matching engines deploy multiple hot backups of a matching engine cluster in a manner of mutual backup, and one matching engine provides service at the same time, when the matching engine fails and cannot continue to work normally, the matching engine cluster can rapidly detect the failure and select one backup matching engine to take over the task of the failed matching engine so as to ensure the normal operation of the whole matching system.
7. A memory algorithm based matching system according to claim 6, the matching engine cluster defines the running matching algorithm as a deterministic state machine in a memory state machine copying mode and deploys a plurality of copies of the deterministic state machine to a plurality of matching engines of the matching engine cluster, the matching algorithm copies on each of the matching engines are run from the same initial state, when the matching engine cluster receives the trade orders sent by the gateway cluster, each matching engine matches the trade orders and generates transaction records in turn, meanwhile, updating the independent state of a deterministic matching algorithm state machine, when the matching engine cluster operates normally, the matching algorithm copies on each matching engine have the same result state, when the matching engine cluster fails or is abnormal, the states of the matching engines are inconsistent, and therefore system abnormality is judged.
8. The matching system based on the memory algorithm as claimed in claim 7, wherein the memory state machine replication mode comprises solving reliable multicast and global ordering of matching engine orders by using atomic multicast, providing fast order matching by using a pipelining matching technology based on lock-free order queues, realizing interaction with a database by using an asynchronous consistency persistence technology, performing state monitoring on a matching engine cluster by using a failover technology and ensuring fault-tolerant capability of the system, and solving recovery of a failed matching engine or addition of a new matching engine by using a progress pursuit technology.
9. The memory algorithm-based matching system as claimed in claim 1, wherein the database builds an index in the order table to improve query efficiency, and the index is a single index.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111622475.0A CN114283001A (en) | 2021-12-28 | 2021-12-28 | Matching system based on memory algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111622475.0A CN114283001A (en) | 2021-12-28 | 2021-12-28 | Matching system based on memory algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114283001A true CN114283001A (en) | 2022-04-05 |
Family
ID=80876854
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111622475.0A Pending CN114283001A (en) | 2021-12-28 | 2021-12-28 | Matching system based on memory algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114283001A (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002007707A (en) * | 2000-06-22 | 2002-01-11 | Keio Gijuku | Transaction system |
| CN107330788A (en) * | 2017-06-15 | 2017-11-07 | 雷盈科技（上海）有限公司 | A kind of block chain digital asset transaction match method, system, device and medium |
| CN109886765A (en) * | 2017-12-06 | 2019-06-14 | 长沙泽影网络科技有限公司 | A kind of automatic match system of Intelligent E-commerce of J2EE platform |
| CN110264353A (en) * | 2019-05-28 | 2019-09-20 | 必成汇(成都)科技有限公司 | High Availabitity trade match system memory-based and method |
| CN110390596A (en) * | 2019-07-30 | 2019-10-29 | 上海金融期货信息技术有限公司 | A kind of financial futures trading system |
| CN111861639A (en) * | 2020-06-28 | 2020-10-30 | 深圳市中农易讯信息技术有限公司 | Financial transaction matching method and system |
-
2021
- 2021-12-28 CN CN202111622475.0A patent/CN114283001A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002007707A (en) * | 2000-06-22 | 2002-01-11 | Keio Gijuku | Transaction system |
| CN107330788A (en) * | 2017-06-15 | 2017-11-07 | 雷盈科技（上海）有限公司 | A kind of block chain digital asset transaction match method, system, device and medium |
| CN109886765A (en) * | 2017-12-06 | 2019-06-14 | 长沙泽影网络科技有限公司 | A kind of automatic match system of Intelligent E-commerce of J2EE platform |
| CN110264353A (en) * | 2019-05-28 | 2019-09-20 | 必成汇(成都)科技有限公司 | High Availabitity trade match system memory-based and method |
| CN110390596A (en) * | 2019-07-30 | 2019-10-29 | 上海金融期货信息技术有限公司 | A kind of financial futures trading system |
| CN111861639A (en) * | 2020-06-28 | 2020-10-30 | 深圳市中农易讯信息技术有限公司 | Financial transaction matching method and system |
Non-Patent Citations (2)
| Title |
|---|
| 孙长昊: ""高可靠高性能金融交易撮合系统设计"", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 黄宪日;黄忠东;叶德伟;: "高性能证券交易撮合系统的设计和实现", 计算机应用与软件 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| JP5536939B2 (en) | Failover system and method | |
| CA2911001C (en) | Failover system and method | |
| US8898669B2 (en) | Methods and systems for coordinated transactions | |
| US20070083521A1 (en) | Routing requests based on synchronization levels | |
| US20240111547A1 (en) | Systems and methods of distributed processing | |
| US20090187600A1 (en) | Method of improving replica server performance and a replica server system | |
| CN114283001A (en) | Matching system based on memory algorithm | |
| US20230305726A1 (en) | Data placement and recovery in the event of partition failures | |
| US11522966B2 (en) | Methods, devices and systems for non-disruptive upgrades to a replicated state machine in a distributed computing environment | |
| US20240104519A1 (en) | Actor model payment processing engine | |
| AU2012202229B2 (en) | Failover system and method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |