AnalyticDB for MySQL is a real-time data warehousing service developed by Alibaba Cloud to process petabytes of data. It is the only service of its kind that has been tried and tested in core businesses on an ultra-large scale.
After its initial release in Alibaba Group in 2012, AnalyticDB for MySQL has so far been iterated through nearly 100 versions and supports online analysis for a variety of businesses owned by Alibaba Group, such as e-commerce, advertising, logistics, entertainment, tourism, and risk control. In 2014, AnalyticDB for MySQL was officially released by Alibaba Cloud. It began to provide services for both traditional large and medium-sized enterprises and public service sectors as well as many Internet companies in more than a dozen industries.
AnalyticDB for MySQL V3.0 is an upgraded cloud-native data warehousing service that is developed based on eight years of experience and integrates database and big data technologies.
AnalyticDB for MySQL V3.0 adopts a cloud-native architecture to separate computing resources from storage resources and hot data from cold data. It supports real-time data write with high throughput, strong data consistency, and hybrid loads of highly concurrent queries and high-throughput batch processing.
The first layer is the access layer and consists of Multi-Master linearly scalable coordination nodes. The access layer is used for protocol layer access, SQL parsing and optimization, real-time sharding of written data, data scheduling, and query scheduling.
The second layer is the computing engine and supports the integrated execution of distributed massively parallel processing (MPP) and directed acyclic graphs (DAGs). The computing engine can work with intelligent optimizers to support hybrid loads of highly concurrent queries and complex SQL statements. Additionally, the cloud-native infrastructure enables compute nodes to be flexibly scaled out within seconds and make the most of resources based on business requirements.
The third layer is the distributed storage engine and supports real-time data write with strong consistency and high availability in compliance with the Raft consensus protocol. The storage engine uses data sharding and Multi-Raft to support parallel processing, tiered storage to separate hot and cold data and reduce costs, and hybrid row-column storage and intelligent indexing to provide extreme performance.
AnalyticDB for MySQL can restore services within seconds, automatically detect faults, remove and rebuild replicas, and support deployment within the same zone or cross zones based on the three-layer architecture. AnalyticDB for MySQL stores data in three replicas and supports full and incremental data backup. This can provide you with the data reliability required in the finance industry. AnalyticDB for MySQL also provides auxiliary tools for data migration, data synchronization, data management, data integration, and data security. This makes AnalyticDB for MySQL easy to use and allows you to focus more on business development.
AnalyticDB for MySQL combines the advantages of distributed architecture, elastic computing, and cloud computing to significantly improve its scalability, usability, reliability, and security and therefore meet the requirements of data warehousing in different scenarios. AnalyticDB for MySQL supports concurrent access on a larger scale, provides faster read and write performance, and implements smarter management of hybrid query loads. It allows you to utilize resources in a finer-grained manner and at a lower cost, which enables you to focus more on business development and data value.
AnalyticDB for MySQL provides a variety of modes and specifications.
- Elastic modeAnalyticDB for MySQL in elastic mode utilizes an architecture that separates storage resources from computing resources. This allows you to separately scale computing resources, data I/O resources, and storage space to improve resource utilization. AnalyticDB for MySQL also supports the tiered storage of hot and cold data, isolation by resource pool, and time-specific scaling features. You are charged based on the actual storage space and data retention period.
- Basic specifications in elastic mode: Specifications that contain 8 cores and 32 GB of memory or 16 cores and 64 GB of memory are basic specifications in elastic mode. These specifications do not ensure high-availability services or support isolation by resource pool or time-specific scaling. Tiered storage of hot and cold data is supported. These specifications are ideal for scenarios where you require accelerated queries of business intelligence (BI) reports.
- Advanced specifications in elastic mode: Specifications that contain 32 cores and 128 GB of memory or more are standard specifications in elastic mode. These specifications can provide high-availability services and support isolation by resource pool, time-specific scaling, and tiered storage of hot and cold data. These specifications are ideal for scenarios where you require accelerated queries of BI reports and real-time data warehousing.
- Reserved mode
AnalyticDB for MySQL in reserved mode allows you to purchase computing resources and storage space of specific specifications. Data I/O resources are bound to computing resources. You are charged based on the specifications that you purchase.
Cluster Edition in reserved mode: C8 and C32 instance types are provided. C8 contains 24 cores and 192 GB of memory, and C32 contains 96 cores and 768 GB of memory. These instance types provide good query performance but generate higher storage costs. You can use these instance types in high concurrency scenarios where less than 10 TB of data is queried to maximize cost-effectiveness.