Informatica PowerCenter Developer Specialist key terms

Central metadata database that stores mappings, sessions, workflows, source and target definitions, and other shared objects. It lets PowerCenter clients and services reuse consistent definitions across projects.

Container inside the PowerCenter Repository that organizes related objects and controls permissions. Developers commonly group mappings, sessions, and workflows by project or subject area.

Repository object saved once and referenced by multiple mappings, mapplets, or workflows. Editing the reusable object can change every object that references it.

Relationship in which one repository object references another, such as a mapping using a source definition or a session using a mapping. Dependency analysis shows impact before changes.

Repository feature that tracks saved versions of objects, allowing developers to compare, retrieve, and manage changes over time. It supports controlled development and rollback.

Repository object describing the structure and metadata of data a mapping reads, including columns, data types, keys, and source database properties. It does not contain the source data.

Repository object describing the table or file a mapping writes, including columns, data types, keys, and load properties. It defines the expected output structure.

Transformation representing rows read from a relational or file source and converting source data types to PowerCenter-native types. It can apply SQL overrides, filters, joins, and sorting.

Passive transformation that calculates new values row by row using expressions, variables, and functions. It can output derived columns, filter flags, or values for later transformations.

Passive transformation that passes only rows meeting a specified condition and discards all others. It is often used to remove unwanted records before further processing.

Active transformation that evaluates multiple conditions and routes each input row to one or more output groups, optionally including a default group for unmatched rows.

Active transformation that groups rows and computes aggregate values such as sums, counts, averages, minimums, and maximums. Sorted input and incremental aggregation can improve performance.

Active transformation that combines rows from two sources using a join condition and supports normal, master, and detail outer joins. It can join heterogeneous sources directly.

Passive or active transformation that retrieves values from a lookup table, view, file, or pipeline and returns them to the current row, often to validate or enrich data.

Passive transformation that creates unique numeric values, commonly used to generate surrogate keys. It can be reusable or non-reusable and can reset based on input values.

Active transformation that sets row-level insert, update, delete, or reject flags so the session applies the intended change to the target.

Active transformation that sorts data by one or more ports in ascending or descending order. Sorted data can improve aggregator and joiner performance and support distinct processing.

Active transformation that converts a single input row with repeating groups into multiple output rows. It is often used for COBOL sources or denormalized data.

Active transformation that selects top or bottom N rows according to rank conditions and grouping ports, similar to SQL ranking functions.

Active transformation that defines commit and rollback points within a mapping by evaluating expressions for each row, overriding the session commit interval.

Transformation that executes SQL statements or stored procedures in a mapping, supporting static and dynamic SQL modes and database-specific syntax.

Reusable set of transformations saved as a repository object and inserted into mappings, with input and output ports defining its interface. It promotes consistency and reduces duplicate design.

Special transformation inside a mapplet that receives data from the calling mapping and exposes ports to the mapplet's internal transformations.

Special transformation inside a mapplet that sends processed data back to the calling mapping through defined output ports for use by downstream transformations.

Named value defined outside a mapping and supplied at run time, allowing the same mapping to use different filters, file paths, or connection values without editing.

Value stored in the repository and changed during session runs, used to carry state across rows or sessions. Its final value can persist for later runs.

Set of linked tasks that the Integration Service executes in a defined order, including sessions, commands, decisions, and event waits. It coordinates the overall data integration process.

Task that executes a mapping, defining source and target connections, load properties, error handling, commit intervals, and performance options. It is the runtime instance of a mapping.

PowerCenter client tool used to create, schedule, monitor, and manage workflows, sessions, and tasks stored in the repository. It also shows run status and logs.

Workflow task that runs an operating system command or script on the Integration Service node, often for file preparation, cleanup, or external job orchestration.

Workflow task that evaluates a condition and controls which downstream path continues based on link conditions, enabling conditional branching in a workflow.

Practice of improving mapping and session throughput by optimizing sources, transformations, caches, partitioning, and database operations. It balances resource use against elapsed time.

Dividing pipeline data into subsets that multiple threads process concurrently, configured at source, target, or transformation level. It can improve throughput but adds complexity and resource use.

Session option that converts mapping logic into SQL and pushes as much processing as possible to the source or target database, reducing data movement.

In-memory or persistent copy of lookup data that avoids repeated database queries. Cache types include static, dynamic, and persistent, each with different refresh behavior.

Number of rows after which a session commits work to the target, balancing transaction size, recovery time, and performance. Smaller intervals can reduce lost work after failure.

Ability to restart a failed session from the point of failure using recovery strategy and commit intervals, avoiding reprocessing all rows.

Session setting that determines how the Integration Service restarts after failure: resume from the last commit point, restart the task, or fail the workflow.

Session feature that captures rows rejected by transformations or targets, along with error messages, into a reject file or error table for analysis.

File written by the Integration Service containing rows that could not be processed, with fields showing the row data and error information.