This topic introduces ApsaraDB for Lindorm (Lindorm) and describes the benefits of ApsaraDB for Cassandra optimized by Lindorm.

Introduction to Lindorm

Lindorm is a cloud-native, multi-model database service that is applicable to all scales. Lindorm provides cost savings in data storage and processing, and supports the pay-as-you-go billing method. Lindorm is compatible with the standard APIs of multiple open source platforms, such as Apache HBase, Apache Cassandra, Apache Phoenix, OpenTSDB, Apache Solr, and SQL. Lindorm uses the following data models to organize and standardize data: time series, wide table, search, and file. Lindorm is intended for enterprises in the Internet, Internet of Things (IoT), Internet of Vehicles (IoV), advertising, social networking, monitoring, gaming, and risk control sectors. Lindorm also provides strong support for business-critical systems of Alibaba Group.

Lindorm uses a cloud-native, multi-model architecture that decouples compute and storage. Lindorm provides the following benefits: elasticity, cost savings, ease of use, high compatibility, and stability. Lindorm allows you to store and analyze the following types of data: metadata, log files, bills, tags, messages, reports, dimension tables, results tables, feeds, user profiles, device data, monitoring data, sensor data, small-sized files, and small-sized images. Lindorm provides the following key features:
  • Multi-model integration: Lindorm supports the following types of data models: wide table, time series, search, and file. Lindorm is compatible with the APIs of multiple open source database management systems, and also provides a separate API. Data can be transformed and synchronized among different models. This makes application development more agile, flexible, and efficient.
  • Cost savings: Lindorm is capable of handling tens of millions of concurrent requests and responds with a latency of a few milliseconds. Lindorm greatly reduces the storage cost by using high-density and cost-effective media, automatic cold and hot data separation, and adaptive compression.
  • Cloud-native support and high scalability: Lindorm allows you to scale computing and storage resources separately. Lindorm also provides serverless services which you can scale and pay for based on actual usage.
  • Open data ecosystem: Lindorm provides multiple easy-to-use features, such as data exchanges, processing, and subscription. In addition, Lindorm is highly compatible with MySQL, Spark, Flink, and Kafka.

Challenges with Apache Cassandra

Apache Cassandra is an open source, distributed NoSQL database that is designed based on Amazon DynamoDB and Google Bigtable. Apache Cassandra is decentralized, supports consistency level customization, and provides a SQL-like language Cassandra Query Language (CQL). However, you may encounter several challenges when you use Apache Cassandra. For example, Apache Cassandra does not decouple compute and storage. If you want to expand storage, you must first migrate data. This complicates resource scaling and makes Apache Cassandra weak in handling traffic spikes. Lindorm uses an architecture that decouples compute and storage to support elastic scaling.

Apache Cassandra requires you to periodically repair all replicas. Otherwise, issues such as ghost keys may occur. However, the repairing task may compromise service stability because it consumes large amounts of system resources. Lindorm uses storage plug-ins to maintain consistency. When you write data to Lindorm, the data is written to a specified number of replicas. This prevents data inconsistency and saves you the hassle of repairing replicas.

Benefits of ApsaraDB for Cassandra optimized by Lindorm

  • Better performance: Compared with Apache Cassandra, ApsaraDB for Cassandra optimized by Lindorm provides higher data throughput and lower latency when handling large amounts of data.
  • Separation of cold and hot data: ApsaraDB for Cassandra optimized by Lindorm uses various storage media and compression algorithms to store cold and hot data separately. This improves the efficiency of retrieving hot data and reduces the expenses on cold data storage.
  • Pay-as-you-go: ApsaraDB for Cassandra optimized by Lindorm provides the Cluster Edition and Serverless Edition. When you use the Serverless Edition, you do not need to manage resources. The Serverless Edition scales resources based on workloads and is billed based on the number of requests and databases. The Serverless Edition supports the pay-as-you-go billing method and is easy to use and cost-effective.
  • Indexes and wide tables: ApsaraDB for Cassandra optimized by Lindorm uses full-text indexes to accelerate queries on wide tables, and supports queries that are written in CQL.
  • Various data channels: ApsaraDB for Cassandra optimized by Lindorm provides various data channels, such as search engines, online transaction processing (OLTP) systems, data warehouses, and log queues.
  • Enterprise-class features: ApsaraDB for Cassandra optimized by Lindorm provides a variety of enterprise-class features. For more information, refer to the following figure. For more information about the enterprise-class features, go to the product page of Lindorm.