ApsaraDB for Lindorm (Lindorm) is a cloud native multi-model database service. It allows you to store data of all sizes. Lindorm supports low-cost storage and processing of large amounts of data and the pay-as-you-go billing method. It is compatible with the open standards of multiple open source software, such as Apache HBase, Apache Cassandra, Apache Phoenix, OpenTSDB, Apache Solr, and SQL. Lindorm supports the following types of models: time series, wide table, search, and file. It is the preferred database choice for scenarios such as the Internet, Internet of Things (IoT), Internet of Vehicles (IoV), advertising, social networking, monitoring, gaming, and risk control. Lindorm also provides strong support for the core business of Alibaba Group.
- Multi-model integration: Lindorm supports four types of models: wide table, time series, search, and file. It provides a unified API and also supports the APIs of multiple open source database management systems. Data can be transferred and synchronized between models. This makes application development more agile, flexible, and efficient.
- Ultimate cost-effectiveness: Lindorm supports tens of millions of concurrent requests and millisecond-level latency. Lindorm greatly reduces the cost of data storage by using high-density and cost-effective medium, automatic cold and hot data separation, and adaptive compression.
- Cloud native and high scalability: Lindorm provides independent auto scaling for computing and storage resources. Lindorm also provides pay-as-you-go serverless services that allow you to perform instant scaling based on your business requirements.
- Open data ecosystem: Lindorm provides multiple simplified management capabilities such as data exchanges, processing, and subscription. It can be seamlessly integrated with systems such as MySQL, Spark, Flink, and Kafka.
Lindorm supports four types of models: wide table, time series, search, and file. It provides a unified API and also supports the APIs of multiple open source database management systems. Data can be transferred and synchronized between models. This makes application development more agile, flexible, and efficient. The core multi-model capabilities are provided by the following four data engines:
- Wide table engine
The wide table engine is used to store large amounts of key-value data and table data. It provides global secondary indexes, multi-dimensional queries, dynamic columns, and Time to Live (TTL). It is applicable to scenarios such as the storage of metadata, orders, bills, user personas, social networking information, feeds, and logs. The wide table engine is compatible with the open standards of multiple open source software, such as Apache HBase, Apache Phoenix, and Apache Cassandra.
The wide table engine supports tens of millions of concurrent requests, separation of cold and hot data, and hundreds of petabytes of data storage. Compared with the performance delivered by open source Apache HBase, the throughput is increased by 2 to 6 times, the percentile 99% (P99) latency is reduced by 90%, the mean time to repair (MTTR) is reduced by 90%, the data compression ratio is increased by 100%, and the comprehensive storage cost is reduced by 50%.
- Time series engine
The time series engine is used to store and process time series data such as measurement data and device operational data in scenarios such as IoT and monitoring. It provides an HTTP API and is compatible with the OpenTSDB API. It also supports SQL queries. The time series engine uses a compression algorithm dedicated to time series data. The data compression ratio can reach up to 15:1. The time series engine supports multi-dimensional queries and aggregate computing of large amounts of data and provides downsampling and pre-aggregation.
- Search engine
The search engine provides capabilities such as full-text searches, aggregate computing, and complex multi-dimensional queries of large amounts of text and document data. It can also be seamlessly used to store the indexes of the wide table and time series engines. This accelerates the data retrieval and queries. The search engine is applicable to scenarios such as the queries of logs, bills, and user personas. It is compatible with the open standards of the open source Solr platform.
- File engine
The file engine supports access to the underlying storage that is shared by the file, wide table, time series, and search engines. This way, the underlying data files can be imported, exported, computed, and analyzed in a more efficient manner. The file engine is compatible with the open standards of the open source Hadoop Distributed File System (HDFS).
Some scenarios, such as monitoring, social networking, and advertising, may use a combination of HBase and Elasticsearch or a combination of HBase, OpenTSDB, and Elasticsearch. In such cases, you may have multiple pain points, such as complex architectures, complex queries, weak consistency, high cost, and misalignment of features. The cloud native multi-model capabilities of Lindorm can help you resolve these pain points to ensure efficient innovations.