CN115563174A - Method and system for managing user hot spot data in memory - Google Patents

Abstract

The present invention provides a method for managing user hotspot data in the internal memory. The method is as follows: 1. Define the structure of the user linked list; 2. When the user logs in, search for the user from the user dictionary in the internal memory, and if it does not exist, search for the user from the database User, then add this user to the head of the first-generation user linked list; 3. When the user enters the game, look up the user from the user dictionary in the memory, add the user to the head of the second-generation user linked list, and delete it from the first-generation user linked list , 4. After the user enters the activity, look up the user from the user dictionary in the memory, add the user to the head of the 3rd generation user linked list, and delete it from the 2nd generation user linked list. Look up the user pointer in the dictionary, read the user data in the user linked list through the pointer, if it does not exist, read it from the database and insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary; to ensure that active users are more likely to be Save it.

Description

Method and system for managing user hotspot data in memory

technical field

The invention relates to the technical field of data storage, in particular to a method and system for managing user hotspot data in memory.

Background technique

The so-called "static hotspot data" refers to hotspot data that can be predicted in advance. For example, it can be screened out in advance through the registration of sellers, and these hot products can be marked through the registration system. In addition, big data analysis can also be used to discover hot products in advance, such as analyzing historical transaction records and user shopping cart records to find out which products may be more popular and sell better. These are hot spots that can be analyzed in advance.

The so-called "dynamic hotspot data" refers to hotspots that cannot be predicted in advance and are temporarily generated during the operation of the system. For example, a seller advertised on Douyin, and then the product became popular immediately, causing it to be bought in large quantities in a short period of time.

In the game server, user information is not only stored in the database and distributed cache, but also maintains a copy of user data in the program memory, because the program memory has no network loss and is the fastest way to read it. However, due to the scarcity of memory, the memory capacity of a server is limited, and it is impossible to cache a full amount of user data in the memory, so how to maintain hot user data as much as possible under the limited memory, this is what this patent wants to solve question.

Contents of the invention

In order to overcome the above-mentioned problems, the object of the present invention is to provide. While not increasing memory capacity, it is more likely that active user data will be preserved.

The present invention adopts the following solutions to realize: a method for managing user hotspot data in memory, said method comprising the following steps:

Step S1, define the structure of the user linked list, divide the user linked list into 3 segments, namely the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, the 3 segment linked lists are connected sequentially from right to left, and use a dictionary to store user numbers and User linked list data pointer;

Step S2, when the user logs in, look up the user from the user dictionary in the internal memory, if it exists, return the user data, and go to step S3; if it does not exist, look up the user from the database, and then add this user to the first-generation user linked list Head, and recorded as the node pointer of the head of the 1st generation user linked list;

Step S3, when the user enters the game, look up the user from the user dictionary in the memory, add the user to the head of the 2nd generation user linked list, delete it from the 1st generation user linked list, and record it as a node pointer at the head of the 2nd generation user linked list;

Step S4, after the user enters the activity, look up the user from the user dictionary in the memory, add the user to the head of the 3rd generation user linked list, delete it from the 2nd generation user linked list, and record it as the node pointer of the 3rd generation user linked list link head;

Step S5, when user data needs to be read, look up the user pointer from the user dictionary in the internal memory, read the user data in the user linked list through the pointer, return the user data if it exists, and insert it into 1 after reading it from the database if it does not exist Replace the head of the user linked list and store the pointer to the user dictionary.

Further, the structure of the user linked list is as follows {uid, pre, next, userinfo, group}, wherein, uid: user number; pre: pointer to the previous user data; next: pointer to the next user data; userinfo: specific User information; group: data generation type, the group value is 1, 2, 3, indicating algebra; all user data are connected through pre and next pointers to form a user linked list, which stores user data just like the memory cache .

Further, the user dictionary stores the user number and the user linked list data pointer, which is defined as: map[int]*user, the keyword key of the int type is the user number, and the value *user type is the user pointer, pointing to the specific user linked list data , so that the position of the data in the linked list can be obtained most quickly through the user uid.

Further, when the user linked list stores data, if the total length of the user linked list is greater than the preset length value, start from the end of the user linked list, move from the 1st generation user linked list to the 3rd generation user linked list, and remove the set amount of user data .

Further, the step S5 is further specifically: when reading the user data, look up the data first from the user dictionary map, if it exists, return the user data, if it does not exist, read the user data from the database and insert it into the first-generation user linked list head, and store pointers to the user dictionary map, where the preset length value is 10000, that is, by default, only the data of 10000 active users in memory is stored. If the total length of the user linked list is greater than 10000, start from the end of the linked list and pass The pre of the user linked list structure advances to the left one by one, and removes the last 100 user elements from the end, so that users with higher algebra and positions further to the left are more likely to be stored in memory data.

The present invention provides a system for managing user hotspot data in memory. The system includes: a linked list definition module, an active user information storage module, an active user information storage module, an active user information storage module, and a data reading module;

The linked list definition module defines the structure of the user linked list, and divides the user linked list into 3 sections, which are respectively the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, and the 3 section linked lists are connected successively from right to left, and are stored in a dictionary User number and user linked list data pointer;

The active user information storage module, when the user logs in, searches the user from the user dictionary in the memory, if it exists, returns the user data, and enters the active second user information storage module; if it does not exist, it searches the user from the database , and then add this user to the head of the first-generation user linked list, and record it as the node pointer of the first-generation user linked list link head;

The active second user information storage module, when the user enters the game, searches the user from the user dictionary in the internal memory, adds the user to the head of the 2nd generation user linked list, and deletes it from the 1st generation user linked list, and records it as the 2nd generation user linked list The node pointer of the chain head;

The active three user information storage module, after the user enters the activity, searches the user from the user dictionary in the memory, adds the user to the head of the 3rd generation user linked list, and deletes it from the 2nd generation user linked list, and records it as the 3rd generation user linked list The node pointer of the chain head;

The data reading module, when user data needs to be read, looks up the user pointer from the user dictionary in the internal memory, reads the user data in the user linked list through the pointer, returns the user data if it exists, and reads it from the database if it does not exist Then insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary.

Further, the structure of the user linked list is as follows {uid, pre, next, userinfo, group}, wherein, uid: user number; pre: pointer to the previous user data; next: pointer to the next user data; userinfo: specific User information; group: data generation type, the group value is 1, 2, 3, indicating algebra; all user data are connected through pre and next pointers to form a user linked list, which stores user data just like the memory cache .

Further, the user dictionary stores the user number and the user linked list data pointer, which is defined as: map[int]*user, the keyword key of the int type is the user number, and the value *user type is the user pointer, pointing to the specific user linked list data , so that the position of the data in the linked list can be obtained most quickly through the user uid.

Further, when the user linked list stores data, if the total length of the user linked list is greater than the preset length value, start from the end of the user linked list, move from the 1st generation user linked list to the 3rd generation user linked list, and remove the set amount of user data .

Further, the implementation of the data reading module is further specifically: when reading user data, first look up data from the user dictionary map, if it exists, return the user data, if it does not exist, read the user data from the database and insert Go to the head of the first-generation user linked list, and store the pointer in the user dictionary map, where the preset length value is 10000, that is, only store the data of 10000 active users in memory by default. If the total length of the user linked list is greater than 10000, then from Starting from the end of the linked list, move forward one by one to the left through the pre of the user linked list structure, and remove the last 100 user elements starting from the end, so that users with higher algebra and positions further to the left are easier to be stored in memory data.

The beneficial effects of the present invention are: using the idea of user generation, users are divided into generations through different business behaviors, and connected by front and back pointers to form a generational linked list, and the linked list forms an activity level from left to right. Decreasing trend, if the memory is insufficient, to remove the user data in the memory, start to remove from the end of the rightmost chain. This ensures that active users are more likely to be saved.

Description of drawings

Fig. 1 is a schematic flow chart of the method of the present invention.

Fig. 2 is a working principle diagram of the system of the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings.

Please refer to shown in Fig. 1, a kind of method of user hotspot data management in memory of the present invention, described method comprises the following steps:

Step S1, define the structure of the user linked list, divide the user linked list into 3 segments, namely the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, the 3 segment linked lists are connected sequentially from right to left, and use a dictionary to store user numbers and User linked list data pointer;

Step S2, when the user logs in, look up the user from the user dictionary in the internal memory, if it exists, return the user data, and go to step S3; if it does not exist, look up the user from the database, and then add this user to the first-generation user linked list Head, and recorded as the node pointer of the head of the 1st generation user linked list;

Step S3, when the user enters the game, look up the user from the user dictionary in the memory, add the user to the head of the 2nd generation user linked list, delete it from the 1st generation user linked list, and record it as a node pointer at the head of the 2nd generation user linked list;

Step S4, after the user enters the activity, look up the user from the user dictionary in the memory, add the user to the head of the 3rd generation user linked list, delete it from the 2nd generation user linked list, and record it as the node pointer of the 3rd generation user linked list link head;

Step S5, when user data needs to be read, look up the user pointer from the user dictionary in the internal memory, read the user data in the user linked list through the pointer, return the user data if it exists, and insert it into 1 after reading it from the database if it does not exist Replace the head of the user linked list and store the pointer to the user dictionary.

The present invention will be further described below in conjunction with a specific embodiment:

A method for managing user hotspot data in memory:

Step 1. Define the structure of the user linked list {uid, pre, next, userinfo, group}, and divide the linked list into 3 segments, namely the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, and the 3 segmented linked list from right to left They are connected in turn, and use a dictionary to store user numbers and user linked list data pointers.

Firstly, it is defined that the user data in the memory is stored in a linked list structure. The user structure is as follows {uid, pre, next, userinfo, group}, uid: user number; pre: pointer to the previous user data; next: next user data pointer; userinfo: specific user information; group: data generation type, the value is 1, 2, 3, indicating algebra. All user data are connected through pre and next up and down pointers to form a user linked list. As a memory cache, the linked list stores user data, but in order to obtain user data as quickly as possible, the user dictionary stores user numbers and user linked list data pointers, and the user dictionary stores user pointers map[int]*user, key of type int It is the user number, and the value *user type is the user pointer, which points to the specific user linked list data, so that the position of the data in the linked list can be obtained most quickly through the user id.

Step 2. When the user logs in, look up the user from the user dictionary in the internal memory. If it does not exist, look up the user from the database, and then add this user to the head of the first-generation user list, and record it as the node pointer of the first-generation user chain head .

When the user logs in, first check whether the id exists in the map, and return the user data directly if it exists. If it does not exist, read the user information userinfo from the database and record it as usern. Assume that the current 1st generation user chain head pointer points to user1a, the pre of user1a points to user2a, and the next of user1a points to user1b. The form of the linked list is as follows: xxx-user2b-user2a-user1a -user1b-xxx, where xxx before and after means ellipsis, that is, there are other data in the front and back; now usern needs to be changed to the chain head of the 1st generation user list, then user2a.next points to usern, usern's pre points to user2a , usern.next points to user1a, and user1a.pre points to usern. Finally, a new linked list is formed: xxx-user2b-user2a-usern-user1a-user1b-xxx, and the latest 1st generation user link head pointer is recorded as usern.

Step 3. When the user enters the game, look up the user from the user dictionary in the memory, add the user to the head of the 2nd generation user linked list, delete it from the 1st generation user linked list, and record it as the node pointer of the 2nd generation user link head.

When a user enters the game, the user is regarded as a second-level active user at this time, and the user whose key is id is searched from the memory dictionary map, and *user points to usern, and the node user1a pointed to by usern.next is recorded as the first-generation user chain head Pointer node, and then move usern to the front of the current user 2 generation chain head user2b to form the following linked list: xxx-usern-user2b-user2a-user1a-user1b-xxx. After usern is moved to the head of the 2nd generation user chain list, usern is recorded as a new 2nd generation user chain head pointer node.

Step 4. After the user enters the activity, look up the user from the user dictionary in the memory, add the user to the head of the 3rd generation user linked list, delete it from the 2nd generation user linked list, and record it as the node pointer of the 3rd generation user link head.

When a user enters an activity, the user is regarded as a third-level active user at this time, and the user whose key is id is searched from the memory dictionary map, and *user points to usern, and the node user2b pointed to by usern.next is recorded as the second-generation user chain head Pointer node, and then move usern to the front of the 3rd generation chain head of the current user to form the following linked list: usern-xxx-user2b-user2a-user1a-user1b-xxx. After moving usern to the head of the 3rd generation user list, record usern as the new 3rd generation user chain head pointer node

Step 5. When the system needs to read user data, look up the user pointer from the user dictionary in the memory, and read the user data in the user linked list through the pointer. If it does not exist, read it from the database and insert it into the head of the first-generation user linked list , and store pointers in the map, where the preset length value is 10000. If the total length of the linked list is greater than 10000, start from the end of the user linked list, move from the first generation of users to the third generation of users, and remove 100 user data.

When the system needs to read user data, it first looks up the data from the user dictionary map. If it exists, it returns the user data. If it does not exist, it reads the user data from the database and inserts it into the head of the first-generation user linked list, and stores the pointer to the map. , the system defaults to only storing the data of 10,000 active users in the memory. If the total length of the linked list is greater than 10,000, start from the end of the linked list, advance to the left one by one through pre, and remove the last 100 user elements from the end, then reach The effect of is that users with higher algebra and positions further to the left are more likely to be stored in memory data.

Please refer to Fig. 2, the present invention provides a system for managing user hotspot data in memory, the system includes: a linked list definition module, an active user information storage module, an active second user information storage module, and an active three user information storage module module, data reading module;

The linked list definition module defines the structure of the user linked list, and divides the user linked list into 3 sections, which are respectively the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, and the 3 section linked lists are connected successively from right to left, and are stored in a dictionary User number and user linked list data pointer; the user linked list structure is as follows {uid, pre, next, userinfo, group}, wherein, uid: user number; pre: the pointer of the previous user data; next: the pointer of the next user data ; userinfo: specific user information; group: data generation type, the group value is 1, 2, 3, indicating algebra; all user data are connected through pre and next pointers to form a user linked list, and the user linked list is the same as the memory cache , which stores user data.

The active user information storage module, when the user logs in, searches the user from the user dictionary in the memory, if it exists, returns the user data, and enters the active second user information storage module; if it does not exist, it searches the user from the database , and then add this user to the head of the first-generation user linked list, and record it as the node pointer of the first-generation user linked list link head;

Described user dictionary stores user number and user linked list data pointer, is defined as: map[int]*user, the keyword key of int type is the user number, and value*user type is the user pointer, points to concrete user linked list data, like this through The user uid can get the position of the data in the linked list most quickly.

The active second user information storage module, when the user enters the game, searches the user from the user dictionary in the internal memory, adds the user to the head of the 2nd generation user linked list, and deletes it from the 1st generation user linked list, and records it as the 2nd generation user linked list The node pointer of the chain head;

The active three user information storage module, after the user enters the activity, searches the user from the user dictionary in the memory, adds the user to the head of the 3rd generation user linked list, and deletes it from the 2nd generation user linked list, and records it as the 3rd generation user linked list The node pointer of the chain head;

The data reading module, when user data needs to be read, looks up the user pointer from the user dictionary in the internal memory, reads the user data in the user linked list through the pointer, returns the user data if it exists, and reads it from the database if it does not exist Then insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary.

When the user linked list stores data, if the total length of the user linked list is greater than the preset length value, start from the end of the user linked list, move from the 1st generation user linked list to the 3rd generation user linked list, and remove the set amount of user data.

The implementation of the data reading module is further specifically: when reading user data, first look up data from the user dictionary map, if it exists, return the user data, if it does not exist, read the user data from the database and insert it into the first generation The head of the user linked list, and store pointers to the user dictionary map, where the preset length value is 10000, that is, only store the data of 10000 active users in memory by default, if the total length of the user linked list is greater than 10000, start from the end of the linked list , move forward one by one to the left through the pre of the user linked list structure, and remove the last 100 user elements starting from the tail, so that users with higher algebra and positions further to the left are more likely to be stored in memory data.

In short, the present invention uses the idea of user generation to divide users into generations through different business behaviors, and connects the front and back pointers to form a generational linked list, and the linked list forms a decreasing trend of activity from left to right. If the memory is insufficient, to remove the user data in the memory, start removing from the end of the rightmost chain. This ensures that active users are more likely to be saved.

The above descriptions are only preferred embodiments of the present invention, and all equivalent changes and modifications made according to the scope of the patent application of the present invention shall fall within the scope of the present invention.

Claims (10)

1. a method for user hotspot data management in internal memory, is characterized in that: described method comprises the steps: Step S1, define the structure of the user linked list, divide the user linked list into 3 segments, namely the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, the 3 segment linked lists are connected sequentially from right to left, and use a dictionary to store user numbers and User linked list data pointer; Step S2, when the user logs in, look up the user from the user dictionary in the internal memory, if it exists, return the user data, and go to step S3; if it does not exist, look up the user from the database, and then add this user to the first-generation user linked list Head, and recorded as the node pointer of the head of the 1st generation user linked list; Step S3, when the user enters the game, look up the user from the user dictionary in the memory, add the user to the head of the 2nd generation user linked list, delete it from the 1st generation user linked list, and record it as a node pointer at the head of the 2nd generation user linked list; Step S4, after the user enters the activity, look up the user from the user dictionary in the memory, add the user to the head of the 3rd generation user linked list, delete it from the 2nd generation user linked list, and record it as the node pointer of the 3rd generation user linked list link head; Step S5, when user data needs to be read, look up the user pointer from the user dictionary in the internal memory, read the user data in the user linked list through the pointer, return the user data if it exists, and insert it into 1 after reading it from the database if it does not exist Replace the head of the user linked list and store the pointer to the user dictionary. 2. the method for user hotspot data management in a kind of internal memory according to claim 1, is characterized in that: described user linked list structure is as follows {uid, pre, next, userinfo, group}, wherein, uid: user number; pre : pointer to the previous user data; next: pointer to the next user data; userinfo: specific user information; group: data generation type, the group value is 1, 2, 3, indicating algebra; all user data is passed through pre and Next, the upper and lower pointers are connected to form a user linked list. The user linked list is the same as the memory cache, storing user data. 3. the method for user hotspot data management in a kind of memory according to claim 2, is characterized in that: described user dictionary stores user number and user linked list data pointer, is defined as: map[int]*user, int type The keyword key is the user number, and the value *user type is the user pointer, pointing to the specific user linked list data, so that the position of the data in the linked list can be obtained quickly through the user uid. 4. The method for managing user hotspot data in a kind of memory according to claim 1, characterized in that: when the user linked list stores data, if the total length of the user linked list is greater than the preset length value, then start from the end of the user linked list, Move from the 1st generation user linked list to the 3rd generation user linked list, and remove the set amount of user data. 5. The method for managing user hotspot data in memory according to claim 4, characterized in that: said step S5 is further specifically: when reading user data, first look up data from the user dictionary map, if it exists, then Return the user data. If it does not exist, read the user data from the database and insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary map. The preset length value is 10000, that is, only 10000 active users are stored by default. For the data in the memory, if the total length of the user linked list is greater than 10000, start from the end of the linked list, advance one by one to the left through the pre of the user linked list structure, and remove the last 100 user elements starting from the tail, so that the higher the algebra, the lower the position Users on the left are more likely to be kept in memory data. 6. A system for user hotspot data management in memory, characterized in that: the system includes: linked list definition module, active one user information storage module, active two user information storage module, active three user information storage module, data reading module; The linked list definition module defines the structure of the user linked list, and divides the user linked list into 3 sections, which are respectively the 1st generation user linked list, the 2nd generation user linked list, the 3rd generation user linked list, and the 3 section linked lists are connected successively from right to left, and are stored in a dictionary User number and user linked list data pointer; The active user information storage module, when the user logs in, searches the user from the user dictionary in the memory, if it exists, returns the user data, and enters the active second user information storage module; if it does not exist, it searches the user from the database , and then add this user to the head of the first-generation user linked list, and record it as the node pointer of the first-generation user linked list link head; The active second user information storage module, when the user enters the game, searches the user from the user dictionary in the internal memory, adds the user to the head of the 2nd generation user linked list, and deletes it from the 1st generation user linked list, and records it as the 2nd generation user linked list The node pointer of the chain head; The active three user information storage module, after the user enters the activity, searches the user from the user dictionary in the memory, adds the user to the head of the 3rd generation user linked list, and deletes it from the 2nd generation user linked list, and records it as the 3rd generation user linked list The node pointer of the chain head; The data reading module, when user data needs to be read, looks up the user pointer from the user dictionary in the internal memory, reads the user data in the user linked list through the pointer, returns the user data if it exists, and reads it from the database if it does not exist Then insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary. 7. The system of user hotspot data management in a kind of memory according to claim 6, characterized in that: the user linked list structure is as follows {uid, pre, next, userinfo, group}, wherein, uid: user number; pre : pointer to the previous user data; next: pointer to the next user data; userinfo: specific user information; group: data generation type, the group value is 1, 2, 3, indicating algebra; all user data is passed through pre and Next, the upper and lower pointers are connected to form a user linked list. The user linked list is the same as the memory cache, storing user data. 8. A system for managing user hotspot data in memory according to claim 7, wherein the user dictionary stores user numbers and user linked list data pointers, and is defined as: map[int]*user, int type The keyword key is the user number, and the value *user type is the user pointer, pointing to the specific user linked list data, so that the position of the data in the linked list can be obtained quickly through the user uid. 9. The system for managing user hotspot data in a kind of memory according to claim 6, characterized in that: when the user linked list stores data, if the total length of the user linked list is greater than the preset length value, starting from the end of the user linked list, Move from the 1st generation user linked list to the 3rd generation user linked list, and remove the set amount of user data. 10. The system for managing user hotspot data in memory according to claim 9, characterized in that: the implementation of the data reading module is further specifically: when reading user data, it is preferentially searched from the user dictionary map Data, if it exists, return the user data, if it does not exist, read the user data from the database and insert it into the head of the first-generation user linked list, and store the pointer in the user dictionary map, where the preset length value is 10000, that is, the default is only Store the data of 10,000 active users in the memory. If the total length of the user linked list is greater than 10,000, start from the end of the linked list, advance one by one to the left through the pre of the user linked list structure, and remove the last 100 user elements from the tail. Algebraic Users who are taller and farther to the left are more likely to be stored in memory data.

Images