MicroStrategy Developer flashcards

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

Semantic layer

A governed metadata layer of reusable attributes, facts, metrics, filters, and prompts that reports and dossiers reference by object ID, so a single definition change propagates to every consumer.

Attribute

A business concept used for grouping and filtering, such as Customer or Region. It holds forms like ID and DESC and maps to a lookup table used in generated SQL joins.

Fact

A numeric column-level object mapped to a specific warehouse table, representing raw data at its stored grain. Facts become metrics only when aggregated by a formula.

Metric

A formula combining facts, attributes, and functions that produces an aggregated value. Its aggregation level is set by the attributes placed on the report unless a level is fixed.

Compound metric

A metric built from other metrics, inheriting their dimensionality and filters. Nesting is limited, so frequently reused nested expressions should be promoted to standalone metrics.

Level metric

A metric whose aggregation is pinned to specified attributes regardless of what the report displays, used with relationship filters to control dimensionality explicitly.

Attribute form

A named representation of an attribute, such as ID or DESC. The ID form drives joins and sorting; the DESC form drives display labels in reports.

Attribute relationship

A declared cardinality (one-to-many, many-to-many, one-to-one) between attributes that determines automatic join paths. Misdeclared relationships cause Cartesian joins or inflated metric totals.

Conditionality

Restricting a metric's input rows before aggregation, applied either as a filter on the metric source or as a level metric with a relationship filter, producing different SQL and results.

Database instance

A connectivity definition holding driver, server, database, and authentication details. Database connections map a metadata project to one or more instances for query execution.

VLDB properties

Settings that control generated SQL behavior, including join types, index hints, temporary table creation, and pre/post statements, configurable at metric, template, or report level.

Intelligent Cube

A multi-dimensional in-memory structure built from a report that answers later queries without returning to the warehouse. It is refreshed on schedule or on demand.

Report cache

Stored result sets keyed by report definition plus user security filters. A cache hit skips warehouse execution but may return stale data until expiry or purge.

Cache invalidation

The process that clears cached results on schema or object changes, scheduled expiry, or manual purge. Until invalidation, users may see results from an earlier warehouse state.

In-memory dataset

A dossier data source held in memory rather than queried live from the warehouse. It returns results until refreshed, so freshness depends on the refresh schedule.

Connection pooling

Reusing a bounded set of database connections across queries. Pool size, query timeouts, and row limits are set per instance to protect the warehouse from overload.

Multi-source reporting

Combining data from multiple database connections in one report, either federated by the warehouse or assembled through in-memory datasets. MicroStrategy does not join across warehouses itself.

Security filter

A predicate appended to every query for a user or group, implementing row-level security. A user in multiple groups receives the union of filters unless intersection is configured.

Object ACL

A per-object access list granting or denying view, execute, edit, delete, control, or grant rights. Deny takes precedence, and rights are inherited from parent folders and groups.

Privilege

A capability granted to users or groups at project or server level, such as creating a dossier. Privileges gate actions independently of any object's access control list.

Project security role

A bundle of privileges such as Project Administrator, Analyst, or Viewer. Custom roles should clone a built-in bundle rather than edit it when the original behavior is still needed.

Audit logging

Recording of connections, logins, report executions, and object changes. Enterprise Manager and audit tables expose usage and performance history, with retention set by configuration.

Authentication methods

Ways users prove identity, including standard credentials, LDAP, Windows, SAML, and Kerberos. Authentication is the first layer before project access, object rights, and security filters.

Object migration

Moving governed objects between environments through project duplication or migration utilities rather than direct production edits, preserving versioned metadata and audit history.

Model Context Protocol

An open protocol exposing tools and resources to AI clients. A MicroStrategy MCP server publishes governed semantic objects as callable tools so models query definitions rather than raw tables.

REST API authentication

Obtaining a token or OAuth credential before calling project, dataset, report, and dossier endpoints. Tokens expire, so clients must refresh before expiry and retry with backoff.

Asynchronous API job

A long-running REST operation that returns a job identifier instead of results. The client polls the job endpoint until completion, then retrieves the output payload.

Embedding SDK

Client tooling that renders dossiers and reports inside another application using an authenticated session token, so embedded content respects the user's security filters.

Semantic-layer AI grounding

Resolving natural-language questions through governed metrics, attributes, and filters so AI answers obey definitions and row-level security instead of generating ungoverned SQL.

Prompt object

A reusable question component that filters a report at run time. In search-driven interfaces, prompt and object metadata help map natural-language input to governed objects.

API pagination

Paged delivery of large result sets through offset or cursor parameters. Clients must follow all pages, since a single response returns only a subset of matching objects or records.

MCP tool surface

The set of semantic objects exposed to an AI client. Keeping it narrow reduces the chance the model selects an unintended metric or attribute during question resolution.