How to protect data security? Here is an RDS disaster recovery plan for you

Today's world is an Internet world full of data, and data is constantly generated in all aspects of life, such as travel records, consumption records, web pages browsed, messages sent, and so on. In addition to text-type data, images, music, and sounds are all data. For enterprises, data is more important production material, and the loss of key data may bring a fatal blow to the enterprise. Once customer data, technical documents, financial accounts and other customer, transaction, and production data are lost or damaged, enterprises may be in trouble.

In the cloud era, cloud databases have a highly reliable and highly available architecture to ensure that user data will not be lost. For example, JD Cloud Cloud Database RDS provides stable and reliable cloud database services based on the world's popular MySQL, Percona, MariaDB, SQL Server, and PostgreSQL database engines.

Compared with traditional databases, ApsaraDB for RDS is easy to deploy, manage and expand. It supports master-slave high-availability architecture by default, and provides a complete set of database solutions such as data backup and recovery, configuration changes, monitoring and alarming, and completely solves the troubles of database usage and operation and maintenance. ApsaraDB for RDS usually provides region-level high availability switching and multi-AZ deployment. However, when various scenarios such as region-level failures and human deletion occur, it is still necessary to design data recovery plans based on business levels to ensure business continuity in all aspects. sturdiness and reliability.

In data disaster recovery solutions, RPO and RTO are two important indicators to measure disaster recovery solutions:

RPO (Recovery Point Objective) is the data recovery point objective, which mainly refers to the amount of data loss that the business system can tolerate;

RTO (Recovery Time Objective) is the recovery time objective, which mainly refers to the longest time the business can tolerate out of service, that is, the shortest time period from the occurrence of the disaster to the recovery of the service function of the business system.

From the figure below, we can see that ApsaraDB for RDS supports RTO and RPO by various disaster recovery related functions:

Function point RTORPO region range automatic backup high middle and high middle same region manual backup medium high same region cross-region backup synchronization medium medium different regions database backup DBS high high different regions

JD Cloud Database Backup DBS

For disaster recovery scenarios, JD Cloud has launched an independent centralized database backup service - database backup DBS.

The database DBS can back up data to the cloud or remote computer room in real time, providing continuous, low-cost data protection for the database, helping enterprises to quickly build a database backup center to meet business and regulatory backup compliance requirements. At the same time, JD Cloud Cloud Database RDS also provides functions of automatic backup, manual backup, and cross-regional backup synchronization, providing a complete and cost-effective disaster recovery solution for user systems.

At present, the database DBS provides the entire database backup and recovery capabilities, and provides solutions for disaster recovery scenarios such as preventing database deletion and running away, and disaster recovery in the computer room.

Database backup DBS extends the target database for backup and recovery to other self-built databases other than RDS instances, and can quickly help users to back up self-built databases and cloud RDS instances off-site, and back up databases to databases in real time through dedicated lines or public networks Backup on DBS.

Advantages and Features of Database Backup DBS

Database backup DBS has the following advantages:

(1) Simple and easy to use

Use database backup DBS for database backup, just a few simple steps to create a backup task and start data backup;

(2) Feature-rich

Provide logical backup, incremental backup and other backup options;

(3) low cost

Provide cold backup and disaster recovery solutions to reduce business and storage costs;

(4) Reliable performance

Automatically expand according to the backup volume and backup pressure to adapt to high-concurrency and large-traffic backup scenarios;

(5) Safe and reliable

Support backup availability and integrity check to ensure backup availability.

Database backup RDS backup method

JD Cloud Cloud Database provides three different backup methods: automatic backup, manual backup, and cross-regional backup.

1. Automatic backup

Automatic backup includes full physical backup and binlog incremental backup.

The automatic backup of cloud data RDS is enabled by default. The RDS service will periodically trigger the backup of the RDS instance according to the backup policy. You can select the time period for automatic backup in the backup policy window, and the system will automatically perform the automatic backup within this time period. Start the backup operation at any point in time.

When the amount of data in the RDS instance reaches a certain order of magnitude with the growth of the business volume, the backup task may not be completed within the specified time period, and the backup task will continue in the next period to ensure that the RDS instance can normally complete a full backup.

After configuring automatic backup, you no longer need to perform manual backup every day. The RDS service will ensure that the backup is performed on time and completely save the backup file to the specified location to avoid the loss of your important data due to failure or abnormal operation.

Every key process such as instance creation, deletion, and capacity expansion will trigger automatic backups, including full backups, binlog backups, and even database configuration files and error logs, ensuring user data security in all aspects.

ApsaraDB for RDS performs a Binlog backup every 5 minutes. Binlog configuration is row format, image=full, records all data operations. In summary, RDS instances can provide minute-level RPO for services, and RTO is determined according to the actual data volume.

2. Manual backup

You can trigger a manual backup at any time according to your business needs. After the manual backup is completed, the backup files will not be automatically cleared. You need to manually delete the backup. When the RDS instance is deleted, the backup files are also automatically deleted.

3. Cross-regional backup

ApsaraDB for RDS instances supports high-availability deployment across Availability Zones. Even if a single AZ (availability zone) cannot provide services normally, ApsaraDB for RDS instances can still provide services normally. However, if the Region (region) where the ApsaraDB for RDS instance is located cannot provide services normally, the business that depends on the ApsaraDB for RDS instance will be paralyzed. Cross-region backup synchronization service provides higher availability for ApsaraDB for RDS and realizes cross-region service disaster recovery.

Cross-region backup synchronization service, by synchronizing the backup files of ApsaraDB for RDS instances to different regions (regions), to ensure that ApsaraDB for RDS instances can quickly synchronize the regions where ApsaraDB for RDS backups are when the original region (region) is unavailable. Create a new ApsaraDB for RDS instance.

ApsaraDB for RDS initiates a cross-region backup task every ten minutes. Since the backup task is an asynchronous operation, and data transmission is also a time-consuming process, although data synchronization between different regions is performed based on dedicated lines, the synchronization speed far exceeds the bandwidth of ordinary public networks. However, the synchronization delay of cross-region backup tasks is closely related to the amount of data, so the actual delay time is determined according to the transmission time of the backup file.

At the same time, database backup RDS backup has been designed with many optimizations:

The RDS instance adopts the physical hot backup method. During the backup process, it does not affect the user's access to the database DML operation;

The backup of the RDS instance is performed on the slave database and will not have any impact on the master database;

RDS instance backup is uploaded to OSS in streaming mode. The backup data will not occupy the storage space of the instance. Data can be uploaded concurrently and at high speed on demand to reduce backup time. 1T data can be backed up in 2-3 hours;

RDS instance backup supports compression and encryption. The compression ratio can reach 3:1, and protect the security of user backup data;

Backup files of RDS instances are stored in object storage. Object storage provides ultra-high data reliability and ensures the security of backup files. And the object storage bucket is hosted by the RDS service. You cannot view the object storage bucket through the console, but the RDS service provides a download link for the backup file, and you can directly download the backup file to the local.

Data Recovery

In the process of using an ApsaraDB for instance, if a software exception or some unexpected situation causes dirty data in the database, you can restore the database in full according to the backup or create a new database instance, or you can choose to restore one or more database tables in the database.

In addition, if your business needs to perform remote disaster recovery, you can select a backup file in the cross-region backup to restore a new database instance in the specified region. RDS uses streaming mode to concurrently download backup files from OSS to local data recovery, and supports concurrent encryption and decryption strategies to further reduce recovery time.

Currently, RDS provides four recovery strategies: recovery based on local overwrite of backups, creation of new instances from basic backups, creation of new instances by point in time, and recovery of local instances based on point in time. Among them, the time of creating and restoring policies by point in time can be accurate to the second level. According to the point-in-time recovery, single-database and single-table restoration can be performed on local data, and local original database tables can be retained. By comparing the same data at different times, the database's The specific operation is convenient for users to perform further data audit on the data before and after recovery.

write at the end

This article focuses on ApsaraDB for RDS to introduce database disaster recovery solutions. In addition to RDS instances, a complete disaster recovery solution should also include hosts, LBs, and other resources for deploying your application. The disaster recovery plan needs to be repeatedly tested and verified in the actual production environment. A complete disaster recovery plan can allow you to discover potential problems in time, so that you can more calmly deal with various accidents and risks in the production environment and discover potential problems in time.