Denodo Associate flashcards

32 cards. Read the term and the definition here. The deck at the top of this page is optional.

Logical data warehouse

An architecture where a virtualization layer sits above existing warehouses, marts, and other sources, exposing one logical schema to consumers while data stays in place.

Base view vs derived view

A base view wraps a single source object and reads it directly; a derived view is built over other views and applies joins, filters, or transformations without changing sources.

Query delegation (pushdown)

The optimizer sends part or all of a query to the source database for native execution, reducing data transferred to the Denodo engine. Non-delegable parts run in the engine.

Virtual vs materialized data

Virtual data is read from the source at query time and is always current; materialized or cached data is a stored copy refreshed on a schedule and can be stale.

Interface view purpose

An interface view publishes a stable contract of columns and semantics to consumers, so the underlying implementation can be replaced without breaking client queries.

Virtual DataPort server

The Denodo component that stores view and data source metadata and executes client queries, translating them into source queries and combining the results.

Design Studio role

The Eclipse-based development client where administrators and developers create data sources, views, and permissions; it is separate from the runtime web tier.

Solution Manager function

A centralized tool for exporting, importing, and promoting artifacts such as views, data sources, and policies between environments like development, test, and production.

Denodo Scheduler

A component that runs scheduled jobs, including cache refreshes, exports, and multi-step flows, with calendars, dependencies, and failure alerts.

Denodo Data Catalog

A web portal for discovering, profiling, and requesting access to published views; it provides business metadata and lineage rather than administrative view editing.

VDP cluster architecture

Multiple Virtual DataPort servers behind a load balancer sharing one metadata repository, with a separate cluster for catalog and web presentation components.

Embedded MPP role

An optional in-memory massively parallel processing engine that holds cached data and answers high-concurrency analytical queries without repeatedly hitting source systems.

Diagnostic and Monitoring tool

A server utility that reports health, query statistics, and logs, and shows running, waiting, and completed queries with their execution details.

Data source creation order

Create the data source with connection details, introspect the schema, create base views from imported tables, then build derived views on top of them.

JDBC driver requirement

Each JDBC source needs the matching driver installed on the VDP server; a missing or mismatched driver prevents connection and introspection.

Connection pooling settings

Minimum and maximum pool size, validation query, and timeout values control how source connections are reused and how dead connections are detected.

Primary key declaration purpose

Declaring a primary key on a base view enables caching, some join optimizations, and correct handling of duplicate rows during aggregation.

Pass-through authentication

A configuration where the client's credentials are forwarded to the source instead of using stored credentials, so the source enforces each end user's identity.

Query wrapping mode

An adapter option that encloses the source query in an outer SELECT to add filters or limits; it affects which operations can be delegated to the source.

Cost-based optimizer

The Denodo component that estimates row counts and costs per branch and chooses join order, join method, and delegation strategy; it depends on current statistics.

Full cache mode

The entire view result is stored and served from cache until invalidated by a time-to-live, a schedule, or an explicit refresh.

Partial cache mode

Query results are cached per query pattern rather than as a whole-view snapshot, so repeated identical or similar queries are served from the cache.

Execution trace

A per-query plan showing each node, the SQL delegated to sources, rows returned, and elapsed time, used to locate the actual bottleneck.

Summary (materialized) view

A precomputed aggregation stored for reuse; the optimizer can rewrite a matching aggregation query to read the summary instead of the detail data.

Cross-source join cost

When joined tables live in different sources, both sides must be fetched and joined in the engine or a temporary table, so transfer volume drives performance.

Row-level security policy

A rule that filters rows by session attributes such as current user or department, applied centrally so every consumer sees only permitted records.

Column masking policy

A rule that obfuscates or partially replaces sensitive column values at query time, so consumers receive masked data without a separate copy.

Data lineage vs impact analysis

Lineage traces where a view's data comes from; impact analysis lists which downstream views and consumers break if a source or view changes.

Data sovereignty via virtualization

Keeping regulated data at its source in its jurisdiction and querying it remotely, so only results, optionally masked, cross the boundary.

REST and OData delivery

Denodo publishes views as REST web services returning XML or JSON, and as OData endpoints, so non-JDBC applications can query the same logical views.

Scheduler export job

A scheduled task that runs a query and writes results to a file or target database, with chaining, calendars, and alerts on failure.

Marketplace access request flow

A consumer discovers a dataset, inspects metadata and samples, requests access, receives owner or role approval, and then connects through the published interface.