Amazon Redshift

Amazon Redshift
Name	Amazon Redshift
Developer	Amazon Web Services
Released	2012
Operating system	Cloud computing
Genre	Data warehouse
License	Proprietary

Amazon Redshift. It is a fully managed, petabyte-scale data warehouse service offered as part of the Amazon Web Services cloud platform. Designed for high-performance analysis using SQL queries, it allows organizations to run complex queries against exabytes of structured data. The service automates administrative tasks like provisioning, setup, and scaling, enabling users to focus on deriving insights from their data.

Overview

Launched by Amazon Web Services in 2012, the service quickly became a cornerstone for analytics in the cloud computing ecosystem. It is built for online analytical processing (OLAP) and business intelligence workloads, contrasting with transactional databases like Amazon Aurora. The underlying technology is based on a massively parallel processing architecture that leverages columnar data storage and advanced compression. This design allows it to deliver fast query performance, even when scanning billions of rows, making it a popular choice for enterprises migrating from traditional on-premises data warehouses like those from Teradata or IBM Netezza.

Architecture

The core architecture employs a leader-node and compute-node structure, coordinating query execution across multiple slices. Data is stored in a columnar format, which enhances compression and speeds up analytical queries by reading only the necessary columns. It utilizes advanced compression encodings and a zone map to minimize I/O. For high availability, data is automatically and continuously backed up to Amazon S3, and features like RAID 0 are used across nodes. The service can also leverage specialized hardware, such as AWS Nitro System instances, to optimize performance and security. Integration with concurrency scaling allows it to handle thousands of concurrent queries by automatically adding transient capacity.

Features

Key capabilities include the ability to scale compute and storage independently, a feature enhanced by the introduction of RA3 instances with managed storage. It supports robust security through integration with AWS Identity and Access Management, Amazon Virtual Private Cloud, and encryption using AWS Key Management Service. For data ingestion, it offers high-speed streaming via Amazon Kinesis and batch loading from Amazon S3. Its performance is boosted by materialized views, result caching, and automatic table optimization. Furthermore, it provides seamless federated querying to run SQL across operational databases like Amazon RDS, Amazon DynamoDB, and data lakes on Amazon S3 via AWS Glue Data Catalog.

Use cases

Organizations deploy it for a wide array of analytical scenarios. A primary use is consolidating data from multiple sources, such as ERP systems, CRM platforms like Salesforce, and transactional databases, into a single source of truth for reporting. It powers business intelligence dashboards and tools from vendors like Tableau Software, Microsoft Power BI, and Qlik. In the realm of log analysis, it is used to query petabytes of application and network logs stored in Amazon S3. The financial services industry utilizes it for risk modeling and fraud detection, while retail companies employ it for analyzing customer behavior and supply chain optimization.

Integration with other services

It is deeply integrated within the broader Amazon Web Services ecosystem. For extract, transform, load (ETL) and data orchestration, it works with AWS Glue and Amazon EMR. Real-time analytics pipelines can be built using Amazon Kinesis Data Firehose for streaming ingestion. Machine learning integration is provided through Amazon SageMaker, allowing data scientists to build models with data directly from the warehouse. For data visualization, native connectors exist for Amazon QuickSight. It also supports querying open-source formats through integrations with the Apache Spark-based AWS Glue and the Presto-powered Amazon Athena.

Pricing model

The service offers several pricing options, primarily based on the type and number of nodes provisioned in a cluster. The traditional model involves on-demand pricing with no long-term commitment, while significant discounts are available through Reserved Instance purchases for predictable workloads. With the RA3 node type, pricing decouples compute and storage, charging separately for managed compute and the actual data stored in Amazon S3. Additional costs may apply for features like concurrency scaling, data transfer across Availability Zones, and backup storage beyond the free allowance. This flexible model competes with other cloud data warehouse solutions like Google BigQuery and Snowflake Inc..

Category:Amazon Web Services Category:Cloud databases Category:Data warehousing