Microsoft Azure Data Fundamentals flashcards

A database optimized for storing and retrieving time-stamped data, like sensor readings or stock prices, efficiently.

• Optimized for time-stamped data
• High ingestion rates
• Efficient time-based queries

Memory trick: Time series data is like a 'timeline' of events.

Aggregation is a data processing operation that combines multiple data points into a single summary value by applying functions like SUM, AVG, COUNT, MIN, or MAX.

• Summarizes data.
• Often used with grouping (e.g., GROUP BY).
• Common in analytical queries and reporting.

Memory trick: Aggregations 'gather' and 'sum' up the data.

ETL is a data integration process that involves extracting data from source systems, transforming it into a desired format, and loading it into a target data store, typically a data warehouse.

• Consists of three main stages: Extract, Transform, Load.
• Used for data warehousing and data integration.
• Prepares data for analytical reporting.

Memory trick: ETL is the 'E'asy 'T'rick for 'L'oading data.

A class of software programs capable of supporting transaction-oriented applications, typically for data entry and retrieval transaction processing.

• Handles high volume of short, atomic transactions
• Optimized for reads and writes
• Emphasizes data integrity and concurrency

Memory trick: OLTP is for 'transactions' happening 'online' right now.

A data processing paradigm where data is processed continuously as it arrives, rather than in batches.

• Handles data in motion, often from IoT devices or clickstreams.
• Enables real-time analytics, anomaly detection, and immediate actions.
• Requires low latency and high throughput.

Memory trick: Streams for Sensors, Not Stagnant Data

OLAP is a category of software tools that provide analysis of data for business intelligence. OLAP systems are optimized for complex data analysis rather than transaction processing.

• Focuses on historical data and trend analysis.
• Optimized for complex queries and aggregations.
• Used for business intelligence, reporting, and forecasting.

Memory trick: OLAP helps you 'look' at the 'past'.

A type of NoSQL database that stores data as a collection of key-value pairs, where a key serves as a unique identifier for its associated value.

• Optimized for fast read/write operations.
• Values can be simple data types or complex objects.
• Does not enforce a schema for values.
• Ideal for session management, user preferences, and caching.

Memory trick: Key-Value pairs are like a fast dictionary for your data.

A NoSQL data model that stores data in flexible, semi-structured documents, often in formats like JSON or XML. Each document is a self-contained unit.

• Offers schema flexibility, allowing documents to have different fields.
• Supports nested data structures.
• Ideal for catalogs, user profiles, and content management.
• Commonly used in Document Databases like MongoDB or Azure Cosmos DB's Document API.

Memory trick: Documents are like flexible folders, holding whatever you put inside.

Organizes data into one or more tables (relations) of rows and columns, with predefined schemas and strong consistency.

• Enforces data integrity through schemas, primary keys, and foreign keys.
• Provides ACID (Atomicity, Consistency, Isolation, Durability) properties.
• Uses SQL for querying and data manipulation.

Memory trick: Relational Ensures Reliability for Records

Data that does not have a predefined data model or is not organized in a pre-defined manner. It accounts for a large majority of data generated today.

• Cannot be stored in traditional relational databases without significant effort.
• Examples include text documents, emails, social media posts, audio, video files.
• Requires advanced analytics techniques (NLP, machine learning) to extract insights.
• Often stored in data lakes or NoSQL document databases.

Memory trick: Unstructured is like a wild, free-flowing river of information.

A NoSQL data model that represents data as nodes (entities) and edges (relationships between entities). It is optimized for storing and querying highly connected data.

• Nodes represent entities (e.g., user, post).
• Edges represent relationships (e.g., 'follows', 'likes').
• Ideal for social networks, recommendation engines, fraud detection.
• Examples include Neo4j, Azure Cosmos DB's Gremlin API.

Memory trick: Graphs show how everything is 'connected'!

A centralized repository that stores a vast amount of raw data in its native format, including structured, semi-structured, and unstructured data.

• Stores data without predefined schema (schema-on-read).
• Highly scalable and cost-effective for large volumes.
• Supports various data processing frameworks (e.g., Spark, Hadoop).
• Used for big data analytics, machine learning, and data exploration.

Memory trick: A Data Lake is a big, raw pool where all your data can swim freely.

Azure Time Series Insights is an end-to-end platform-as-a-service (PaaS) for collecting, processing, storing, querying, and visualizing time series data at IoT scale.

• Optimized for IoT and time-series data.
• Enables near real-time and historical analysis.
• Provides rich visualization and query capabilities over time-stamped data.

Memory trick: Time Series Insights helps you 'see' the 'time' and 'stream' it.

A method of processing data in large groups (batches) at scheduled intervals, rather than in real-time.

• Processes data in chunks
• Scheduled execution
• Suitable for large volumes, non-real-time needs

Memory trick: Batch processing is like baking a 'batch' of cookies once a day.

NoSQL databases provide a mechanism for storage and retrieval of data that is modeled in means other than the tabular relations used in relational databases.

• Handles semi-structured and unstructured data.
• Offers flexible schemas.
• Provides high scalability and availability.

Memory trick: No-SQL is no problem for flexible data.

A type of NoSQL data model that stores data as a collection of key-value pairs, where each key is unique and maps to a single value.

• Optimized for fast read and write operations.
• Values can be simple data types or complex objects.
• Commonly used for caching, session management, and real-time data.

Memory trick: Keys Unlock Value: Choose the right model for your data's journey.

A highly scalable data streaming platform and event ingestion service that can receive and process millions of events per second.

• Designed for high-throughput, low-latency data ingestion.
• Ideal for IoT telemetry, application logging, clickstreams.
• Acts as a 'front door' for data pipelines, buffering events.
• Integrates with stream processing engines like Azure Stream Analytics.

Memory trick: Event Hubs is the 'express lane' for all your streaming data.

A storage tier in Azure Blob Storage optimized for rarely accessed data, offering the lowest storage costs with flexible latency requirements (hours for retrieval).

• Lowest cost storage option for Azure Blob Storage.
• Ideal for long-term backup, archival, and regulatory compliance data.
• Data retrieval can take several hours.
• Supports various data types (structured, semi-structured, unstructured).

Memory trick: Archive is like putting data in a deep, cold freezer for safekeeping.

A cloud search service that gives developers APIs and tools for adding a rich search experience to their applications.

• Provides full-text search, faceted navigation, and geospatial search.
• Can index various data sources, including structured data, Blob Storage (for PDFs), and databases.
• Integrates AI capabilities for enriching content, such as entity recognition and sentiment analysis.

Memory trick: Cognitive Search for Content and Context

Azure Database for PostgreSQL is a managed relational database service in the cloud, based on the open-source PostgreSQL database engine.

• Supports structured data and relational models.
• Enforces primary and foreign key constraints.
• Provides ACID transactional consistency.

Memory trick: PostgreSQL is 'post'-perfect for structured data.

A serverless query service in Azure Synapse Analytics that enables users to analyze data directly in Azure Data Lake Storage Gen2 using T-SQL.

• Queries data without moving or transforming it.
• Pay-per-query model based on data processed.
• Supports various file formats like Parquet, CSV, and JSON.

Memory trick: Serverless queries explore the data lake.

A business intelligence service that provides interactive visualizations and business intelligence capabilities with an interface simple enough for end-users to create their own reports and dashboards.

• Connects to a wide range of data sources.
• Enables creation of interactive reports and dashboards.
• Supports data exploration, drill-down, and automatic refreshes.

Memory trick: Power BI gives you the power to see your data.

A database optimized for storing and analyzing data points that are indexed by time, commonly used for monitoring, sensor data, and IoT applications.

• Designed for data with a timestamp component.
• Efficient for high-volume inserts and time-based queries.
• Supports aggregation and analysis over time windows.

Memory trick: Time-travelers always check the clock.

A massively parallel processing (MPP) data warehousing service in Azure Synapse Analytics, optimized for large-scale analytical workloads.

• Uses columnar storage for efficient analytical queries.
• Scales compute and storage independently.
• Ideal for enterprise data warehousing and business intelligence.

Memory trick: Synapse's SQL pool is the warehouse for big insights.

A feature of Azure Synapse Dedicated SQL pools that allows users to pause compute resources when not in use, significantly reducing operational costs.

• Pausing stops compute billing while retaining data.
• Resuming restarts compute, typically within minutes.
• Ideal for cost optimization during non-business hours or inactive periods.

Memory trick: Dedicated pool, dedicated savings: pause the power!

A cloud-based ETL and data integration service that allows you to create data-driven workflows for orchestrating data movement and transforming data at scale.

• Supports connections to hundreds of data sources.
• Provides visual tools for pipeline development.
• Enables scheduling and monitoring of data integration processes.

Memory trick: Data Factory 'builds' your data pipelines.

A real-time analytics service that processes high volumes of streaming data from various sources to detect patterns and relationships.

• Designed for real-time data processing and analysis.
• Supports SQL-like query language for stream processing.
• Ideal for IoT, log analytics, and clickstream analysis.

Memory trick: Streams are analyzed as they flow.

A highly cost-effective storage tier within Azure Data Lake Storage Gen2, designed for rarely accessed, long-term data retention.

• Lowest cost storage option for ADLS Gen2.
• Optimized for data that can tolerate longer retrieval times.
• Ideal for compliance, audit, and long-term backup data.

Memory trick: Archive your old data to save big bucks over time.

The process of combining data from different sources into a unified view, often involving cleaning, transformation, and standardization.

• Essential for creating a holistic view from disparate datasets.
• Involves ETL (Extract, Transform, Load) or ELT processes.
• Prepares data for analytics, reporting, and machine learning.

Memory trick: Integration brings all the data pieces together into one picture.

A unified analytics service that brings together enterprise data warehousing, big data analytics, and data integration into a single platform.

• Combines SQL technologies for data warehousing with Spark technologies for big data.
• Offers a unified experience for data ingestion, exploration, preparation, transformation, management, and serving.
• Supports both serverless and dedicated resource models for different workloads.

Memory trick: Synapse Unites all Analytics, a true Data Hub.

A cloud-based ETL (Extract, Transform, Load) and ELT (Extract, Load, Transform) service for orchestrating and automating data movement and transformation.

• Used for building complex data pipelines.
• Supports various data sources and destinations.
• Enables data integration and transformation at scale.

Memory trick: The data factory builds the data path.

A real-time analytics service designed to process high volumes of streaming data with low latency.

• Processes data in motion from sources like Event Hubs and IoT Hub.
• Supports SQL-like queries for complex event processing.
• Enables real-time dashboards, alerts, and data archiving.

Memory trick: Stream Analytics flows like a river, processing data as it comes.

A highly scalable and cost-effective data lake solution built on Azure Blob Storage, optimized for big data analytics workloads.

• Supports petabytes of data and trillions of files.
• Offers hierarchical namespace for folder and file organization.
• Combines benefits of Blob Storage (cost-effectiveness, tiered storage) with Data Lake Store Gen1 (file system semantics).

Memory trick: Think of a lake for all your data, big and small.

A unified, Apache Spark-based analytics platform optimized for Azure, providing collaborative workspaces for data engineering, data science, and machine learning.

• Managed Apache Spark clusters for big data processing.
• Supports Python, Scala, R, SQL for data science and ML.
• Ideal for large-scale ETL, stream processing, and advanced analytics.

Memory trick: Databricks is where data scientists get to 'spark' their models.

A serverless query service within Azure Synapse Analytics that enables querying data directly in data lakes using T-SQL.

• No infrastructure to set up or manage.
• Pay-per-query model, ideal for ad-hoc analysis.
• Supports various file formats like Parquet, CSV, JSON.

Memory trick: Serverless SQL queries the lake without needing a boat.