[AWS][Database] Aurora hands-on experiment & comparison of RDS Mysql performance

The experiment includes:

Create an Aurora instance
Use MySQL Workbench to connect Aurora and RDS MySQL
Load data to Aurora and RDS MySQL through dump file
Use query statements to verify the performance of Aurora and RDS MySQL

Task1: Create Aurora database:

To create an Aurora database, just like creating an RDS, log in to the AWS management console, search for Aurora, and create a database.

Database type selection, Aurora with MySQL compatibility, other defaults are fine.

Templates select Dev/Test

DB instance size chooses db.t3.small, because it is a test environment, if it is a production environment, choose Memory Optimized type R series. Local tyrants can also choose high-end examples during the test.

Pay attention to the ports released in the security group. Due to the testing environment, we can release them all.

Keep other configurations as default and create a database.

Then create a RDS mysql database of the same type (db.t3.smal).

After the creation is successful, it is as follows:

Task2: Connect to the springboard and install MySQL workbench:

The springboard can be windows or Linux. I will take windows as an example here:

Task3: Connect Aurora and RDS MySQL through a springboard

Aurora endpoint：

Mysql endpoint：

Task4: Import the SQL Dump file into the database.

For students who are not familiar with SQL dump, this file probably looks like the following figure. It is easy to understand, define the Schema, and then insert the data.

Run Powershell and download the dump file to the desktop

Invoke-WebRequest https://s3-us-west-2.amazonaws.com/aws-tc-largeobjects/SPLs/sharedDatabases/world.sql -OutFile c:\Users\Administrator\Desktop\world.sql

Import Data:

Similarly, Mysql does it again. I won't go into details.

Task5: execute query:

Aurora execution result:

MySQL execution result:

Do you see the difference?
Yes, there is no difference, and the execution speed is very fast. why?
Because there are only 1000 pieces of data, it is difficult to compare the execution efficiency of Aurora and MySQL with a small amount of data. But in the case of millions or tens of millions of data using stress testing tools, their TPS and QPS will have a big gap.

Finally, the comparison result using the pressure measurement tool Sysbench is attached. (I will add it when I think about it)

Detailed experiment video link: https://edu.51cto.com/center/course/lesson/index?id=558304