Skip to main content

Terms

This article summarizes product, task, permission, data security, AI, and OpenAPI terms that appear across the NineData documentation.

A​

  • AccessKey: The access credential ID used when calling NineData OpenAPI. It is usually paired with SecretKey for authentication and request signature verification.
  • AI Model: A model service used by NineData AI capabilities such as SQL generation, diagnosis, optimization, and question answering. Built-in models can be used, and custom models are supported in applicable editions.
  • AI SQL Completion: An AI-assisted SQL Window capability that provides completion suggestions based on the current SQL context.
  • AI SQL Optimization: An AI capability that analyzes SQL statements and provides optimization suggestions, often used for slow SQL analysis, index suggestions, and rewrite ideas.
  • Alert Policy: A configuration that defines alert trigger conditions, notification methods, and recipients for gateways, tasks, or operations monitoring.
  • Alert Receiver Group: A group of users or Webhook channels that receive alert notifications.
  • Assessment Report: An output file generated by a migration assessment task. It records object compatibility, SQL compatibility, risk levels, and workload replay results.
  • Approval Process: A workflow configuration that defines how SQL tasks, permission applications, data import/export tasks, and data archiving tasks are approved.
  • Audit Log: A log that records who performed which operation on which object and when. It is used to trace console activity and security events.

B​

  • Backup and Restore: NineData's data protection module. It supports logical backup, physical backup, online backup data query, and data recovery.
  • Backup Set: A data collection generated by a backup task. It can be used for later restore or online query.
  • Bidirectional Replication: A replication mode in which two data sources synchronize incremental data with each other. It is commonly used for multi-active and disaster recovery scenarios.
  • Breakpoint Resume: A reliability mechanism that lets long-running tasks continue from the completed position after an interruption.

C​

  • Chat2SQL: An AI capability that generates SQL from natural language in SQL Window. It is also referred to as Text2SQL.
  • ChatDBA: NineData's intelligent Q&A assistant. It supports database knowledge Q&A, knowledge base retrieval, SQL troubleshooting, and performance diagnosis.
  • Change SQL: SQL statements used to perform database changes or repair data and structure differences.
  • Cloud Vendor Access Credential: Authorization information used by NineData to access cloud vendor resources. It can be configured with an Access Key or a role.
  • Compatibility: In migration assessment, the result that indicates whether source database objects or SQL can be used directly, used after adaptation, or cannot be used in the target database.
  • Conflict Handling Strategy: The strategy used when replication or archiving encounters existing target data, primary key conflicts, or similar situations, such as ignoring, replacing, or stopping.
  • Connection Address: The host address and port used to access a data source. It may be a public address, private address, internal IP address, or gateway proxy address.
  • Cross-Cloud: A scenario in which business data or tasks flow across multiple cloud vendors, regions, or data centers.

D​

  • Data Archiving and Cleanup: A Database DevOps capability that moves historical data to an archive database or deletes expired data. It can run once or on a schedule.
  • Data Backup: The capability to save database objects or physical files as backup sets for disaster recovery, restore, online query, and data protection.
  • Data Comparison: A module that checks data or structure consistency between two data sources, detects differences, and can generate repair SQL.
  • Data Export/Import: A SQL development capability that exports database data to files or imports external data into a database.
  • Data Replication: A module that synchronizes structure and data between source and target data sources. It supports structure replication, full data replication, incremental data replication, unidirectional replication, and bidirectional replication.
  • Database Group: A logical group that combines multiple databases by rules for unified query, change, or approval management.
  • Database and Table Sharding: An architecture that splits business data across multiple databases or tables. NineData can help manage it with database groups, table groups, and routing algorithms.
  • Data Source: A database, data warehouse, NoSQL, Kafka, Elasticsearch, or similar instance managed in NineData. It usually includes connection address, authentication information, environment, permissions, and access method.
  • Data Source Permission: A user or role's permission to query, modify, replicate, back up, compare, or otherwise operate on a data source.
  • Data Tracking and Rollback: A capability that parses database change logs, locates executed change statements, and generates rollback SQL.
  • Data Consistency: The state in which source and target structures, data content, or business semantics are consistent, often verified by data comparison after replication.
  • Database DevOps: NineData's database development and management module. It provides SQL Window, SQL tasks, rule precheck, approval process, data import/export, slow query analysis, data tracking and rollback, and related capabilities.
  • DataHub: A data source or target type that can be used in certain data replication links.
  • DDL: Data Definition Language, used to create, modify, or delete database objects, such as CREATE, ALTER, and DROP.
  • Development Environment: A data source environment used for development, testing, or validation. It can have permissions, standards, and approval processes that differ from production.
  • Dedicated Cluster: A service cluster composed of servers provided by the user. It executes SQL Window, SQL tasks, data replication, data comparison, backup, and restore workloads.
  • DML: Data Manipulation Language, used to insert, update, or delete data, such as INSERT, UPDATE, and DELETE.

E​

  • Environment: A usage label for a data source, such as production, development, or a custom environment. It can affect permissions, standards, and approval processes.
  • ETL/ELT: ETL (Extract, Transform, Load) is a data extraction, transformation, and loading flow; ELT (Extract, Load, Transform) loads data before transformation. Some NineData replication links support field expressions or ETL functions to transform, clean, and format synchronized data.
  • Execution Plan: The execution path and cost estimate generated by a database for a SQL statement. It is commonly used for query performance analysis.

F​

  • Full Data Replication: A replication stage that synchronizes existing data from the source data source to the target data source.
  • Full Instance Replication: A replication object selection mode that replicates all supported objects in the source data source.

G​

  • Gateway: A database connection service that enables remote access to private network databases and securely connects third-party cloud or local databases to NineData.
  • Geo-Distributed Disaster Recovery: A high availability architecture that continuously synchronizes data to a remote disaster recovery environment and switches to it when the primary region fails.
  • Geo-Distributed Multi-Active: An architecture in which business nodes in multiple regions serve traffic at the same time and keep data synchronized through replication.
  • Group Expression: An expression used in database or table grouping to match database names or table names.

H​

  • Heterogeneous Data Sources: Source and target data sources with different database types. They usually require structure conversion, data type mapping, and SQL compatibility checks.
  • Hybrid Cloud: A business architecture that includes both self-managed databases and cloud databases.
  • Homogeneous Data Sources: Source and target data sources with the same or highly compatible database types.

I​

  • IdP: Identity Provider. In an SSO scenario, it authenticates users and sends identity assertions to the service provider.
  • Incremental Data Replication: A replication stage that continuously synchronizes new or changed data from the source data source, usually based on database logs or change records.
  • In-Site Message: A channel for receiving approval, task, and notification messages inside the NineData console.

K​

  • Knowledge Base: An enterprise knowledge source used to enhance ChatDBA answers. Documents can be uploaded and enabled for retrieval.

L​

  • Logical Backup: A backup method that uses database objects as backup units, such as databases, tables, indexes, and records.

M​

  • Metadata: Information that describes database structures, such as database names, schemas, tables, columns, indexes, constraints, views, functions, and triggers.
  • Migration Assessment: A capability that evaluates object compatibility, SQL compatibility, and required transformation effort before database migration.
  • Multi-Active Marker: A globally unique marker used in multi-source real-time synchronization scenarios to identify data origin and prevent circular replication.
  • Naming Rule: A rule used during structure replication or structure conversion to decide target object name case or whether to keep source naming.

O​

  • OpenAPI: NineData's open interface service. Developers can use HTTP requests to manage resources such as data sources, tasks, roles, and permissions.
  • Organization Mode: An account mode in NineData that supports organization-level user, role, permission, and resource management for team collaboration.

P​

  • Permission: The operation scope available to a user or role in NineData, including management permissions, module permissions, data source permissions, and other permissions.
  • Permission Application: A request submitted when a user lacks required permissions. The user receives access after the request is approved.
  • Personal Mode: An account mode in NineData for individual developers to manage their own data sources and tasks.
  • Physical Backup: A backup method that uses database physical files as backup units.
  • Precheck: An automated check stage before a task is executed. It validates permissions, connectivity, configuration, objects, rules, and compatibility.
  • Private Network Connection: A private network channel between NineData and a target cloud vendor's internal network, avoiding public internet data transmission.
  • Production Environment: A data source environment that carries production workloads and usually requires stricter permissions, rules, and approval processes.
  • Public Network Connection: An access method that connects to a target data source through the internet. It usually requires adding NineData service IP addresses to a database or server allowlist.

R​

  • Real-Time Synchronization: A replication mode that continuously captures incremental source changes and synchronizes them to the target.
  • Regular Account: A user registered through a regular method such as phone number or email. It can log in through regular login or SSO when configured.
  • Reverse Replication: In bidirectional replication, the incremental replication direction from the target data source back to the source data source.
  • Risk Level: A migration assessment indicator that highlights potential risks in object compatibility, SQL compatibility, or replay results.
  • Role: A collection of permissions that controls which modules, data sources, and management operations a user can access.
  • Rollback SQL: SQL statements used to undo or repair the result of a data change. SQL tasks and data tracking and rollback may record or generate rollback SQL.
  • Routing Algorithm: An algorithm used in database and table grouping to locate the target database or table based on field values.
  • RPO: Recovery Point Objective. It measures the acceptable data loss time window in a recovery scenario.

S​

  • SaaS Mode: A ready-to-use cloud service mode in which users can use NineData capabilities without deploying NineData services themselves.
  • Schema: A namespace or object collection in a database. Its meaning and hierarchy vary by database product.
  • SecretKey: The key used to generate OpenAPI request signatures. It must be kept secure and must not be exposed.
  • Sensitive Column: A column marked as sensitive data. Users without authorization cannot view its complete content.
  • Sensitive Data: Data that requires special protection, such as phone numbers, ID numbers, and bank card numbers. NineData supports identification, classification, masking, and permission control for sensitive data.
  • Sensitive Data Level: The security level of sensitive data. NineData provides S0 to S5 by default, where a larger number indicates a higher security level.
  • Sensitive Data Type: A type definition used to identify sensitive columns. It is usually associated with a sensitive data level, identification rule, and masking algorithm.
  • Sensitive Data Masking Algorithm: A rule used to hide, replace, or encrypt sensitive column values so unauthorized users cannot see complete sensitive information.
  • Slow Query: A SQL statement with long execution time or high resource consumption, usually analyzed through slow query analysis, execution plans, or SQL optimization.
  • Source Data Source: The source end that provides data, structure, or SQL in tasks such as replication, comparison, and migration assessment.
  • SP: Service Provider. In an SSO scenario, NineData acts as the SP and uses assertions returned by the IdP to complete login.
  • SSH Tunnel: An access method that connects to a data source through an SSH tunnel, often used when a database must be reached through a bastion host or restricted network.
  • SQL Development Standards: A set of rules used to govern SQL writing and execution behavior. These rules can be checked in SQL Window and SQL tasks.
  • SQL Execution History: Historical records of SQL statements executed by the current account in SQL Window, including execution time, data source, and result.
  • SQL Intelligent Hint: A SQL Window capability that suggests keywords, functions, databases, tables, columns, SQL templates, and syntax help while you type.
  • SQL Review: A capability that checks SQL statements against rules, identifies risks, or provides optimization suggestions.
  • SQL Task: A task that submits database change operations, mainly DDL and DML statements. It supports submission, precheck, approval, execution, and rollback information recording.
  • SQL Window: A database development interface in NineData for online query, SQL execution, result set editing, object viewing, and AI-assisted operations.
  • SSO: Single Sign-On. Users can log in to NineData through a unified identity provider without repeatedly entering credentials in multiple systems.
  • SSO Account: An account created by an SSO organization administrator or automatically joined through SSO. It is usually authenticated and managed by the identity provider.
  • SSO Organization Mode: An organization collaboration mode entered through SSO login. Its permission and resource management behavior is consistent with organization mode.
  • Structure Comparison: A comparison of object definitions such as tables, columns, indexes, constraints, views, functions, and triggers between two data sources.
  • Structure Conversion: A capability that converts table structures between heterogeneous data sources into DDL executable by the target database.
  • Structure Replication: A stage in a data replication task that synchronizes database and table structures from the source data source to the target data source.

T​

  • Table Group: A logical group that combines multiple physical tables by business or sharding rules for unified query, change, or approval management.
  • Target Data Source: The target end that receives data, structure, or validation execution in tasks such as replication, comparison, structure conversion, or migration assessment.
  • Task Lifecycle: The full status flow of a task, from creation, precheck, approval, and execution to completion or failure.
  • Text2SQL: An AI capability that generates SQL from natural language. In SQL Window, it is also referred to as Chat2SQL.
  • Trigger: A database object that automatically executes logic when a specific table operation occurs. It appears in SQL Window, structure comparison, and structure replication scenarios.

U​

  • User-Managed Data Source: A database instance deployed or managed by the user. It can be connected to NineData through public network access, gateway, private network connection, or SSH Tunnel.
  • Unidirectional Replication: A replication mode in which data is synchronized from the source data source to the target data source.

V​

  • View: A database object that acts as a virtual table defined by a query. It can be managed in SQL Window, structure comparison, and structure replication.
  • Visual Table Design: A SQL Window capability for creating or editing table structures through a visual interface.

W​

  • Webhook: A way to send messages or alerts through HTTP callbacks. It can be used for alert receiving and message subscription scenarios.
  • Workload Replay: In migration assessment, the process of sending collected SQL to the target database and using real execution results to validate compatibility.
Last updated on