ApsaraDB for Lindorm (Lindorm) is a cloud native multi-model database service. It allows you to store data of all sizes. Lindorm uses a cloud native architecture in which computing is decoupled from storage. It provides multiple benefits, such as elasticity, cost-effectiveness, stability and reliability, ease of use, and ecosystem compatibility.

Cloud native and high scalability

  • Lindorm uses a distributed architecture in which computing is decoupled from storage. This enables independent auto scaling for computing resources and storage resources.
  • You can scale storage resources online within seconds and computing resources within minutes. The computing resources include the wide table engine, the time series engine, and the search engine.
  • Lindorm provides pay-as-you-go serverless services that enable instant on-demand scaling. Lindorm also supports adaptive scaling and does not require you to manually manage the storage capacity.

Cost-effectiveness

  • Lindorm provides the performance, standard, and capacity storage types to satisfy the needs for cost-effectiveness in different scenarios.
  • Multiple data engines share a unified storage pool. This helps decrease the number of fragments and reduce costs.
  • The unit price of capacity storage is the lowest in the industry. It is significantly lower than that of the self-managed systems that are built on Elastic Compute Service (ECS) instances equipped with local disks.
  • Lindorm provides a built-in optimized compression algorithm. The data compression ratio can reach more than 10:1, which is more than 50% higher than Snappy compression.
  • Lindorm provides built-in adaptive encoding for various data types. Data can be queried without being decoded.
  • Lindorm provides intelligent separation of hot and cold data and allows you to use a mix of storage types. This significantly reduces the overall cost of data storage.

High performance

  • Wide table engine: supports tens of millions of concurrent requests and hundreds of petabytes of storage. The throughput is 3 to 7 times that of open source HBase. The percentile 99% (P99) latency is 10% of that of open source HBase.
  • Time series engine: The write performance and the query performance are 1.3 times those of InfluxDB and 5 to 10 times those of OpenTSDB.
  • Search engine: an optimized version of the Lucene engine. The comprehensive performance is 30% higher than that of open source Solr or Elasticsearch.

Multi-model integration

  • Lindorm can transfer and synchronize data between models. The search engine can be seamlessly used to store the indexes of the wide table engine and the time series engine to accelerate the execution of complex queries.
  • Lindorm allows you to use SQL to interact with your Lindorm instances and run associated queries across multiple data engines.

Ease of use

  • Lindorm is compatible with the open standards of a variety of open source software, such as Apache HBase, Apache Cassandra, Apache Phoenix, OpenTSDB, and Apache Solr. This allows you to seamlessly migrate data to Lindorm.
  • Lindorm provides cloud-based managed services and is free of operations and maintenance (O&M).
  • Lindorm provides a graphical user interface (GUI) for simplified operations. You can use the GUI to manage system settings and access data.

High availability

  • Lindorm uses a distributed multi-replica architecture to implement automatic disaster recovery of clusters. Lindorm ensures a service uptime of more than 99.9%.
  • Lindorm supports cross-zone disaster recovery, ensures strong consistency, and provides high availability capabilities that can meet financial requirements.
  • Lindorm provides active geo-redundancy.

High reliability

  • The underlying layer of Lindorm stores data in multiple replicas to ensure a data reliability of up to 99.99999999%.
  • Lindorm provides enterprise-grade backup capabilities.
  • Lindorm has tens of thousands of nodes deployed on Alibaba Cloud. Lindorm has stood the test of Tmall Double 11 for 10 years.

Ecosystem compatibility

  • Lindorm allows you to migrate data from database management systems such as MySQL, HBase, and Cassandra to Alibaba Cloud without downtime.
  • Lindorm can be seamlessly integrated with computing engines, such as Spark, Flink, Data Lake Analytics (DLA), and MaxCompute.
  • Lindorm supports seamless subscription to and processing of data from log channels, such as Kafka and Log Service.
  • Lindorm provides Lindorm Streams to allow you to obtain incremental data changes in real time. You can subscribe to data changes based on your business requirements.