Cloudera CDP Generalist flashcards

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

CDP Public Cloud control plane

The Cloudera-managed layer that provisions environments, Data Lakes, and Data Hubs; the customer operates the clusters but not the control plane itself.

CDP Private Cloud Base

The on-premises CDP form factor deployed with Cloudera Manager and parcels on customer hosts, providing the full Hadoop ecosystem without the Cloudera-managed control plane.

Shared Data Experience (SDX)

The centralized security and governance layer in CDP combining Ranger policies, Atlas metadata and lineage, and identity, so access rules follow data across clusters and engines.

Data Lake in CDP

The shared foundation holding object storage plus Hive Metastore, Ranger, Atlas, and Knox; Data Hubs and data services attach to it and inherit its policies and metadata.

Data Hub cluster

An ephemeral compute cluster in CDP Public Cloud created from a cluster definition and template; it attaches to a Data Lake and is deleted without affecting the lake's data.

Cloudera Manager parcel

A versioned binary distribution of CDP software that is downloaded, distributed to hosts, and activated; parcels are the supported deployment method and allow side-by-side versions.

Ranger policy evaluation

Ranger evaluates resource and tag policies for each request; an explicit deny overrides an allow, and the default is deny when no policy matches.

Atlas lineage

A graph of relationships between metadata entities showing how data flows from inputs to outputs, captured by Hive, Spark, and Impala hooks during query execution.

Ranger tag-based policy

A policy that references an Atlas classification such as PII; it applies to every resource carrying that tag, so tagging an asset in Atlas automatically subjects it to the policy.

Ranger column masking

A policy that replaces column values at query time with null, redact, or a custom expression; stored data is unchanged, so masking is safe for shared datasets.

Ranger row filtering

A policy that appends a filter expression to queries for matching users or groups; Hive and Impala honor the filter, and the user does not see it in their SQL.

Knox Gateway

A reverse proxy and perimeter security layer for Hadoop REST APIs and UIs; in CDP Public Cloud it authenticates via the identity provider and authorizes via Ranger.

HDFS NameNode HA

High availability for HDFS using an active and standby NameNode, JournalNodes for edit log sharing, and ZooKeeper Failover Controller for automatic failover.

YARN Capacity Scheduler

The default YARN scheduler in CDP; queues have capacity, maximum capacity, and user limits, and preemption can be configured per queue.

Impala COMPUTE STATS

A command that gathers table and column statistics for the Impala optimizer; stale or missing statistics cause poor join orders and broadcast decisions.

Hive on Tez

The Hive execution engine that runs queries as Tez DAGs, grouping small tasks and sizing reducers from input size, replacing MapReduce execution.

Apache Ozone

A scalable object store in CDP with buckets and keys, managed by the Ozone Manager and Storage Container Manager; it offers an S3-compatible API and can replace HDFS for some workloads.

Apache Iceberg tables

A table format in CDP supporting ACID transactions, time travel, and schema evolution; the catalog can be the Hive Metastore or a REST catalog.

Cloudera Data Warehouse

A CDP service providing isolated Impala or Hive virtual warehouses that share Data Lake storage and SDX policies, with auto-suspend and auto-scale.

Cloudera Data Engineering

A CDP service running Spark jobs on Kubernetes virtual clusters, with job dependencies, schedules, Airflow integration, and a CLI for automation.

Cloudera Machine Learning

A CDP service providing projects, Jupyter and RStudio sessions, and model endpoints; projects are Git-backed and governed by SDX policies.

Cloudera DataFlow

A CDP service for NiFi flow-based ingestion and Flink stream processing; flows are authored in NiFi and deployed as CDF deployments.

Cloudera Operational Database

A CDP service providing HBase and Phoenix clusters with a gateway, backup to object storage, and region server autoscaling.

Cloudera Data Visualization

A CDP service for building dashboards connected to CDW, CDE, and CML data sources; access is governed by the underlying data policies.

Replication Manager

A CDP component that replicates HDFS, Hive, and Ozone data between clusters using snapshot-based HDFS replication and Hive Replication for external tables.

CDP CLI

The command-line interface for the CDP control plane; it authenticates with an API access key and manages environments, Data Lakes, Data Hubs, and data services.

Cloudera Resource Name (CRN)

The unique identifier for a CDP resource; it is returned by the CLI and API and is used in IAM policies and API calls to scope permissions.

CDP cross-account IAM role

The AWS role assumed by the CDP control plane to create and manage resources; its trust policy references the Cloudera control plane account and it grants least-privilege permissions.

Data Hub autoscaling

Load-based or schedule-based scaling of Data Hub worker nodes; load-based uses YARN metrics, and scale-down drains nodes before removal.

CDW virtual warehouse

An isolated Impala or Hive compute cluster in Cloudera Data Warehouse; it can be resized, auto-scaled, and auto-suspended while sharing Data Lake storage.

Hive Metastore

The shared catalog storing database, table, column, and partition metadata and storage locations; in CDP Public Cloud it lives in the Data Lake and is shared by all engines.

CDP Private Cloud Data Services

The Kubernetes-based CDP form factor running data services on OpenShift or embedded Kubernetes, managed by the CDP control plane with namespaces and environments.