Data service (OneService) is the final step in building a data mid-end with Dataphin. It acts as a unified access point for data services and provides centralized management of these services. This approach lowers the barrier to data access while ensuring data security.
Prerequisites
You have purchased the data service value-added service and activated the data service module for your tenant.
Feature description
The Dataphin data source feature has been deprecated. You can use the Dataphin Java Database Connectivity (JDBC) Driver as an alternative. For more information, see Connect to Dataphin using JDBC.
Common data application issues
The standard process from requirement submission to delivery includes the following stages: requirement submission, analysis, product design, documentation, development, functional testing, deployment, and delivery. This process often takes about two weeks and requires multiple iterations of development and testing. The resulting interfaces are highly specific and typically meet only the needs of individual business analyses. After an interface is developed, you cannot monitor its usage online. If business analysis requirements change, you must submit a new requirement and schedule another development iteration. This process leads to several issues in data applications:
Long data development cycles and slow responses: Development requires scheduled iterations, and the high barrier to data use prevents immediate application.
Siloed development: This approach has high development costs and low reusability because a specific data interface is created for each requirement.
Poor data security and query performance: Development often prioritizes data accessibility over data resource security and data access stability.
High maintenance costs: Without online platform management, you cannot monitor for exceptions or receive alerts. Online business issues are addressed only after they are discovered, which requires code troubleshooting, fixing, and redeployment.
Value of Data service
To address these common issues, Data service provides two modes for developers: a codeless UI and a flexible code editor. These modes lower the development barrier, improve efficiency, standardize code quality, and simplify change management. For business users, Data service provides secure and stable access to data resources and prevents sensitive data from being exposed at the application layer. Data service also supports service monitoring and alerting. This allows development and operations and maintenance (O&M) teams to promptly identify and resolve issues based on abnormal data or alert notifications. The value of Data service includes the following features:
Data service creation features:
You can associate one or more physical tables to form a logical table. You can then create Application Programming Interfaces (APIs) based on this logical table.
You can create APIs in a codeless UI or a code editor. The codeless UI is suitable for simple queries, while the code editor lets you write custom SQL scripts for complex queries that involve table joins, complex conditions, or complex data processing logic.
Data service management and O&M features:
It provides environment isolation for draft, development, and production. Tests in the development environment use development data, while tests in the production environment use production data.
Provides service management features such as authentication and asset owner changes. These features allow you to manage service invocation requests, data query authentication, and asset owner transfers.
You can configure alerts for service timeouts, failures, and rate limits. You can receive alert notifications through DingTalk chatbots, email, text messages, or phone calls.
You can monitor statistics on abnormal service invocations to view services and applications with abnormal invocation patterns and trends.
You can query detailed service invocation logs to view error messages and the specific SQL statements that were executed.
Advantages of Data service
Unified interface standards: Data service provides a consistent data interface standard and service metadata. This standardizes interface development, reduces the integration effort for downstream applications, and improves data access efficiency.
Security and compliance: Data service reduces the storage and exposure of detailed and sensitive data at the application layer. It uses unified platform security control policies, such as API invocation requests, authentication, flow control, and whitelists, to lower data security management costs.
Service Monitoring: Data service supports statistics on data service invocations and detailed invocation log queries. This helps O&M and development teams to efficiently troubleshoot and resolve anomalies in service invocations.
You can use the Data service platform to create data service interfaces, invoke services, and manage the entire service lifecycle. This improves efficiency because simple configurations on the platform can automatically generate and deploy data services.
Features
With Data service, you can create service project groups based on physical tables from business data sources or logical tables generated by Dataphin. This capability meets various business needs. Applications can then invoke APIs. Data service also supports API O&M and monitoring to help you promptly identify and resolve exceptions during API invocation.
For information about the data source types that Data service can read from physical tables, see Data sources supported by Dataphin.
Scenarios
Data service can be used to build an enterprise API ecosystem. For example, you can open APIs to partners and developers to help your enterprise turn data into valuable assets.
Scenarios and procedures
Before you use Data service, you should understand its scenarios and the operational procedures for each role.