Alibaba Cloud Lindorm is a cloud-native multi-model hyper-converged database service developed and optimized for IoT, the Internet, and the Internet of Vehicles (IoV). Lindorm offers unified capabilities for database access and integrated processing for various data types, including wide tables, time series, files, objects, streams, and spaces. Lindorm is compatible with the standard APIs of various open-source software and services, such as SQL, Apache HBase, Apache Cassandra, Amazon Simple Storage Service (Amazon S3), Apache Phoenix, OpenTSDB, Hadoop Distributed File System (HDFS), Apache Solr, and Apache Kafka. It can also seamlessly integrate with third-party ecosystem tools. Lindorm is suitable for scenarios like log data processing, monitoring, billing data processing, advertising, social networking, travel, and risk management. It also offers robust support for the core business of Alibaba Group.
Lindorm utilizes a cloud-native multi-model architecture where computing is decoupled from storage. It brings benefits such as elasticity, cost-effectiveness, ease of use, high compatibility, and stability. It enables storing and analyzing various data types like metadata, logs, bills, tags, messages, reports, dimension tables, result tables, feeds, user personas, device data, monitoring data, sensor data, small files, and small pictures. Lindorm provides the following core capabilities.
- Multi-model integration: Lindorm supports multiple data models, including wide tables, time series, objects, files, queues, and spaces. Data can be transferred and synchronized between models. Lindorm offers unified and integrated capabilities and services for data access, storage, retrieval, computing, and analysis, enhancing application development agility, flexibility, and efficiency.
- High cost-effectiveness: Lindorm can handle tens of millions of concurrent requests with response latency in milliseconds. It supports multi-level media for data storage and features automatic cold and hot data separation and adaptive compression, reducing storage costs.
- Cloud-native elasticity: Lindorm provides separate autoscaling for computing and storage resources.
- High compatibility: Lindorm is compatible with standard APIs of various open-source software and services, such as SQL, Apache HBase, Apache Cassandra, Amazon S3, Apache Phoenix, OpenTSDB, HDFS, Apache Solr, and Apache Kafka. It seamlessly integrates with Hadoop, Spark, Flink, and Kafka systems, offering easy-to-use features for data transfer, processing, and subscription.
Multi-model capabilities
Lindorm boasts support for a diverse array of data models, encompassing wide tables, time series, objects, files, queues, and spaces. It adheres to standard SQL statements and interfaces with multiple open-source systems' APIs. This facilitates enhanced agility, flexibility, and efficiency in application development. Key multi-model capabilities are driven by the following data engines.
- Wide table engine: The wide table engine serves to manage wide table data and object data, offering features such as global secondary indexes, multi-dimensional queries, dynamic columns, and Time to Live (TTL) functionality. It finds utility in scenarios involving metadata storage, order and billing management, user personas, social networking data, feeds, and logs. The wide table engine aligns with open-source software and services' APIs, including SQL, Apache HBase, Apache Cassandra, and Amazon S3. Capable of handling tens of millions of concurrent requests and storing hundreds of petabytes of data, the wide table engine also incorporates hot and cold data separation. Compared to the performance of open-source Apache HBase, it achieves throughput increases of 2 to 6 times, reduces the 99th percentile (P99) latency by 90%, slashes mean time to repair (MTTR) by 90%, boosts data compression ratios by 100%, and cuts comprehensive storage costs by 50%.
- Time series engine: This engine manages time series data, such as measurement, monitoring, and operational data from industrial devices in sectors like IoT and monitoring. It supports SQL operations for managing, writing, and querying time series data, utilizing a dedicated compression algorithm tailored for such data types. Achieving compression ratios of up to 15:1, the time series engine enables multi-dimensional queries, aggregate computing of large data volumes, downsampling, and continuous queries.
- Search engine: Accelerating the retrieval and analysis of multi-model data, the search engine leverages core technologies like columnar storage and inverted indexing. It furnishes functionalities such as full-text indexing, aggregate computing, and complex multi-dimensional queries, catering to log, billing, and user persona queries. Compliant with standard APIs like SQL and Apache Solr, the search engine is a versatile tool for data exploration.
- File engine: Handling data directories and files, the file engine facilitates efficient access to shared underlying storage utilized by the wide table, time series, and search engines. This enables streamlined importing, exporting, computation, and analysis of underlying data files. Compatibility with the standard HDFS API ensures seamless integration.
- Compute Engine: Integrated with the Lindorm storage engine, the compute engine delivers distributed computing services under a cloud-native architecture, catering to various needs such as data production, interactive analytics, machine learning, and graph computing. Compliant with the standard Apache Spark API, the compute engine ensures compatibility and interoperability.
- Stream computing engine: This engine stores and processes streaming data, offering lightweight computing capabilities. It serves to store streaming data in Lindorm, meeting requirements for streaming data processing and application.
Conclusion
Lindorm stands as a robust and versatile solution, offering a rich array of multi-model capabilities tailored to meet the diverse needs of modern data-driven applications. With support for a wide range of data models, including wide tables, time series, objects, files, queues, and spaces, Lindorm provides a comprehensive platform for managing and processing data efficiently. Whether it's the wide table engine's prowess in handling massive volumes of data with enhanced performance, the time series engine's specialized compression algorithms, or the search engine's ability to accelerate data retrieval and analysis, Lindorm offers a suite of tools designed to empower developers and businesses alike. Furthermore, the seamless integration with standard APIs and compatibility with open-source software and services ensure ease of adoption and interoperability. With Lindorm, organizations can confidently tackle the complexities of modern data management, enabling them to derive valuable insights and drive innovation in their respective domains.
Join the conversation! Your thoughts help the community grow.