Lindorm Distributed Processing System (LDPS) is a high-performance, cost-effective, stable, and reliable distributed computing service that is provided based on the core features of Lindorm. LDPS can meet the computing requirements in business that are implemented based on Lindorm, such as data production, interactive analytics, machine learning, and graph computing. You can use LDPS to process large amounts of data at high concurrency. You are the owner of the LDPS resources in your Lindorm instances. LDPS supports the computing models and programming interfaces of Apache Spark and integrates the features of the storage engine LindormStore. This way, LDPS can fully use the features of underlying data storage and indexing capabilities to efficiently execute distributed computing tasks.
Features
LDPS provides the following features:
Multiple access methods: You can perform interactive data analytics by using Java Database Connectivity (JDBC). You can also submit JAR files to customize distributed computing tasks.
Out-of-the-box: A unified permission system is supported for LDPS and the storage engine LindormStore, including the wide table engine (LindormTable), the time series engine (LindormTSDB), and the search engine (LindormSearch). Therefore, you do not need to configure the complex underlying components of Lindorm. Developers who understand SQL and have Spark development experience can easily use LDPS.
O&M-free: You need only to focus on job management in the console and SparkUI without the need to pay attention to cluster O&M operations, such as configuration, upgrade or downgrade of configurations, and scale-out or scale-in operations.
Auto scaling: LDPS supports auto scaling. This helps you reduce computing costs, save computing resources, and respond to traffic peaks in business with predicated workload fluctuations.
Pay-as-you-go: You are charged only for the computing resources that are actually used.