This topic describes the features and specifications of AnalyticDB for MySQL editions.
AnalyticDB for MySQL editions
AnalyticDB for MySQL launches Data Warehouse Edition in reserved mode, Data Warehouse Edition in elastic mode, Data Lakehouse Edition, Enterprise Edition, and Basic Edition over five years of evolution. The following figure shows the evolution of AnalyticDB for MySQL.
Data Warehouse Edition in reserved mode uses a storage-compute coupled architecture to meet high-performance query requirements. However, Data Warehouse Edition in reserved mode does not allow you to separately add computing resources or storage resources.
To add computing resources during peak hours to ensure query performance and to remove computing resources during off-peak hours to reduce resource costs, AnalyticDB for MySQL launches Data Warehouse Edition in elastic mode that uses a storage-compute decoupled architecture. Data Warehouse Edition in elastic mode supports elastic scaling of computing resources and uses resource groups to physically isolate computing resources to meet multi-tenancy and hybrid workload requirements.
Due to the increase in the demand for analysis of semi-structured and unstructured data, such as log data, AnalyticDB for MySQL launches Data Lakehouse Edition based on elastic mode. Data Lakehouse Edition provides the Spark compute engine to perform batch processing on large amounts of data.
The newly launched Enterprise Edition and Basic Edition integrate the benefits of Data Warehouse Edition in reserved mode, Data Warehouse Edition in elastic mode, and Data Lakehouse Edition. Enterprise Edition and Basic Edition use a storage-compute coupled architecture for reserved resources to ensure expected query performance. In scenarios that involve periodic or sudden changes in business workloads, you can add or remove computing resources based on your business requirements.
Enterprise Edition
Enterprise Edition is an integrated edition of Data Lakehouse Edition and Data Warehouse Edition that provides the same features as Data Lakehouse Edition. Enterprise Edition supports capabilities in elastic mode, such as resource group isolation, elastic resource scaling, and tiered storage of hot and cold data. Enterprise Edition also supports capabilities in reserved mode, such as high throughput, real-time writes and high-concurrency, real-time queries.
Enterprise Edition runs in multi-replica mode and uses a multi-replica storage architecture. Enterprise Edition provides distributed capabilities and high availability and is suitable for the development, testing, and production scenarios of various types of enterprises.
Basic Edition
Basic Edition runs in single-replica mode and provides the same features as Enterprise Edition. Basic Edition uses a single-replica storage architecture and does not support high availability. Basic Edition is suitable for business scenarios that require low-cost hot data storage but do not require high availability.
You cannot change a cluster from Basic Edition to Enterprise Edition.
Data Lakehouse Edition
Data Lakehouse Edition uses a storage-compute decoupled architecture and integrates cost-effective batch processing and high-performance real-time analysis capabilities. Compared with Data Warehouse Edition in elastic mode, Data Lakehouse Edition features fully enhanced data collection, storage, computing, and application capabilities. A single copy of data stored at the underlying storage layer is used to perform both batch processing and real-time analysis. This prevents consistency and timeliness issues that may occur during data synchronization. Computing resources in Data Lakehouse Edition are physically isolated for batch processing and real-time analysis. Data Lakehouse Edition can perform scheduled or automatic scaling of compute and storage resources. Data Lakehouse Edition supports the Spark multi-language programmable compute engine that uses standardized APIs and can be used to batch process large amounts of data. In addition, Data Lakehouse Edition allows you to configure real-time data synchronization to Hudi tables on Object Storage Service (OSS) or C-Store tables in a visualized manner.
Data Lakehouse Edition runs in cluster mode and is ideal for the following scenarios: data processing (such as data cleansing and standardization), multi-source aggregate analysis, wide table development, and prediction and insights (such as machine learning and AI).
Data Warehouse Edition
Elastic mode
Data Warehouse Edition in elastic mode is built on a storage-compute decoupled architecture, and can write large amounts of data in real time and perform high-performance real-time analysis.
Data Warehouse Edition in elastic mode physically isolates computing resources for batch processing and real-time analysis and allows you to individually scale up computing and storage resources based on your business requirements. Data Warehouse Edition in elastic mode also provides tiered storage of hot and cold data, reducing storage costs.
Data Warehouse Edition in elastic mode is ideal for writing large amounts of data in real time, performing complex extract, transform, load (ETL) operations, performing complex queries on large amounts of data, and analyzing historical data and logs.
Data Warehouse Edition in elastic mode is available in Standalone Edition and Cluster Edition.
Standalone Edition
Standalone Edition is deployed on a single node. It does not have the benefits of a distributed architecture or support high availability. Standalone Edition supports tiered storage of hot and cold data. However, it does not support resource group isolation or scheduled scaling. Alibaba Cloud does not provide a service-level agreement (SLA) guarantee for Standalone Edition, and 4 to 8 hours are required for a failover. We recommend that you do not use Standalone Edition in production environments. Standalone Edition is suitable for scenarios that do not require large amounts of data, high queries per second (QPS), or high availability. It is ideal for individual developers to run tests, and startups and small enterprises to handle basic business.
Cluster Edition
Cluster Edition is deployed across multiple nodes, and therefore delivers the benefits of a distributed architecture and supports high availability. Cluster Edition provides more powerful features to help enterprises with their development, testing, and production.
Reserved mode
Data Warehouse Edition in reserved mode is built on a storage-compute coupled architecture. Data Warehouse Edition in reserved mode provides high-throughput real-time writes, high concurrency, and quick response. Data Warehouse Edition in reserved mode is suitable for scenarios such as query acceleration, user profiling, interactive reports, and real-time data services.
Features
The following table compares the features of Enterprise Edition, Basic Edition, Data Lakehouse Edition, Data Warehouse Edition in elastic mode, and Data Warehouse Edition in reserved mode.
Category | Feature | Enterprise Edition | Basic Edition | Data Lakehouse Edition | Data Warehouse Edition in elastic mode | Data Warehouse Edition in reserved mode |
Computing | XIHE analytical compute engine | Supported | Supported | Supported | Supported | Supported |
Spark programmable compute engine | Supported | Supported | Supported | Not supported | Not supported | |
Storage | XUANWU analytical storage engine | Supported | Supported | Supported | Supported | Supported |
Cost-effective Hudi storage | Supported | Supported | Supported | Not supported | Not supported | |
Resource management | Resource group management | Supported | Supported | Supported | Supported only for Cluster Edition | Not supported |
Scheduled scaling | Supported | Supported | Supported | Supported only for Cluster Edition | Not supported | |
Auto scaling | Supported | Supported | Supported | Not supported | Not supported | |
Tiered storage for hot and cold data | Tiered storage for hot and cold data | Supported | Supported | Supported | Supported | Not supported |
Data ingestion | Real-time data import | Supported | Supported | Supported | Supported | Not supported |
Automatic metadata discovery | Supported | Supported | Supported | Not supported | Not supported | |
Job development | SQL job development | Supported | Supported | Supported | Not supported | Not supported |
Spark job development | Supported | Supported | Supported | Not supported | Not supported |
Specifications
Enterprise Edition and Basic Edition
Category | Specifications |
Single-node specifications of reserved resources |
|
Data Lakehouse Edition
Category | Specifications |
Reserved computing resources | Minimum: 16 ACUs Maximum: 4,096 ACUs |
Reserved storage resources | Minimum: 24 ACUs Maximum: 2,064 ACUs |
If you want to purchase more than 512 ACUs of reserved computing resources or more than 256 ACUs of reserved storage resources, submit a ticket.
Data Warehouse Edition in elastic mode
Category | Specifications | |
Computing resources | Standalone Edition | 8 cores and 32 GB memory, 16 cores and 64 GB memory |
Cluster Edition | 32 cores and 128 GB memory or more | |
Storage resources | Elastic I/O resources |
|
Data Warehouse Edition in reserved mode
Model | Specifications | ||
CPU | Memory (GB) | Storage (GB) | |
C8 | 24 cores | 192 | Minimum: 100 Maximum: 2,000 |
C32 | 96 cores | 768 | Minimum: 100 Maximum: 8,000 |
FAQ
How can I view the edition of an AnalyticDB for MySQL cluster?
Log on to the AnalyticDB for MySQL console and go to the Cluster Information page of a cluster. In the Cluster Attributes section, you can view the edition and mode of the cluster.
In what scenarios may availability of an AnalyticDB for MySQL cluster be affected?
The availability of an AnalyticDB for MySQL cluster may be affected when failures occur on the cluster or when the cluster undergoes configuration changes or version updates.