Denodo Platform 9 Certified Developer Professional key terms

A stored connection configuration to an external system such as a relational database, file, or web service. It holds driver, URL, credentials, and pooling settings, and is the starting point for creating base views.

A read-only view created directly over a data source that exposes the source's schema as a Virtual DataPort view. It is the physical layer of the virtual model and is queried through delegation.

A set of reusable, pre-opened connections maintained per data source so queries avoid connection setup cost. Pool size, timeout, and validation settings are configured on the data source.

A data source type that connects to relational databases through a JDBC driver, requiring a driver class, connection URL, and credentials. It supports schema introspection to auto-generate base views.

A view defined by a query over other views rather than a physical table. It can join, filter, aggregate, and union base views, and the optimizer rewrites it into source queries at execution time.

The optimizer's ability to push parts of a query, such as filters, projections, joins, and aggregations, down to the underlying data source so the source does the work instead of the server.

The Virtual DataPort component that transforms a user query on views into an execution plan, deciding which operations are delegated to sources and which run in the server engine.

Per-source declarations of which SQL features, such as joins, subqueries, or specific functions, a source can execute. They constrain what the optimizer may delegate.

A view operation combining rows from two views on a condition. The optimizer may delegate the join to one source or perform it in the server when sources differ.

A derived view that concatenates the result sets of two or more compatible views, optionally keeping duplicates. Column names and types must align, and it is often used to consolidate sources.

A diagnostic representation of how a query was executed, showing delegated source queries and server-side operations. Traces help identify missing delegation and slow steps.

A metadata relationship defined between two views, with a name, cardinality, and join condition. Associations document the model and can drive automatic join suggestions and lineage views.

A label attached to views or columns for classification, such as PII or confidentiality level. Tags support search, governance reporting, and can drive masking policies.

The repository of technical and business descriptions of views, columns, associations, and tags. It powers search, lineage, and documentation exposed to consumers.

A record of how a view or column derives from upstream views and sources. Lineage supports impact analysis before changes and helps trace data quality issues to their origin.

The process of verifying a user's identity, using methods such as database-backed users, LDAP, Kerberos, or SAML. It establishes the session identity used by authorization rules.

The rules that determine what an authenticated user may do, including which views and columns are visible and which operations are permitted. It is enforced by privileges and policies.

A named collection of privileges assigned to users or groups. Roles simplify administration by granting access to many views at once and by supporting inheritance.

A specific permission on a resource, such as connect, execute, or metadata access, granted to a user or role. Privileges are checked at query time.

A rule that obscures column values for unauthorized users, for example by replacing characters or returning null. It is applied at query time so stored data stays unchanged.

A condition attached to a view that restricts which rows a user can see, expressed with the user's session context. It enforces row-level security without duplicating views.

A published HTTP endpoint exposing a view or stored procedure as a resource. It supports JSON and XML payloads, query parameters, and configurable authentication.

A published endpoint using XML envelopes and a WSDL contract to expose views or procedures. Clients generate stubs from the WSDL and invoke operations with typed messages.

A published endpoint following the OData protocol, exposing views as entity sets with standard query options such as filter, select, and expand. It suits tools expecting OData clients.

A published endpoint where clients request exactly the fields they need in a single query. It is generated from the virtual model and supports nested retrieval across related views.

A server-side store of query results that serves repeated identical queries without hitting sources. Modes include full, partial, and incremental caching with configurable expiration.

Techniques that reduce response time, including caching, materialized tables, summary views, and delegation tuning. The goal is to move work to the cheapest capable layer.

A view whose results are precomputed and stored, such as in a database or the server's own storage, then refreshed on a schedule. It speeds expensive aggregations and joins.

A configured unit of scheduled work, such as running a query or exporting data, with a trigger, a source query, and a destination. Jobs can be chained and monitored.

A load that transfers only new or changed rows since the last run, using a timestamp or key watermark. It reduces volume and is configured in the job's source query.

The transfer of query results from Virtual DataPort into a target system, such as a database table or file, executed by a Scheduler job. It may be full or incremental.

The browser-based development environment for building and browsing views, associations, and policies. It provides search, query panels, and model editing for developers.

The component that packages elements from one environment into a revision, exports it, and deploys it to another environment with validation and impact checks. It supports promotion across Dev, Test, and Prod.

The desktop client for administering servers, creating data sources and views, managing users and privileges, and running queries. It offers full administrative control beyond the browser UI.

The layered design in which clients connect to a Virtual DataPort server that parses queries, optimizes them, delegates to sources, and combines results. The server holds the metadata database.

The internal store holding view definitions, users, privileges, and configuration for a Virtual DataPort server. Its backup and restore are essential for disaster recovery.

A deployment of multiple Virtual DataPort servers sharing configuration and metadata to provide failover and load balancing. It requires a shared metadata store and health monitoring.

The steps and entitlement files required to install and start Denodo components, including setting ports, memory, and the metadata database connection. Licensing limits features and scale.