SQreamDB is a columnar analytical SQL database management system designed for large-scale scans, aggregations, and complex queries. It uses GPU parallelism to process data for data warehousing, data preparation, and business intelligence workloads, including datasets ranging from tens to hundreds of terabytes and larger deployments.

Key Capabilities

GPU-accelerated SQL: SQreamDB uses GPU cores to divide and execute large analytical operations in parallel. Queries use substantial support for ANSI SQL and can run through SQream Acceleration Studio or third-party business intelligence tools.
Columnar analytics: Tables are stored by column, reducing unnecessary data reads for analytical workloads. Automatic vertical and horizontal partitioning supports query processing across large datasets.
Data warehousing: Teams can store enterprise-scale data and analyze it through SQL clients, business intelligence tools, and other analytics applications.
Data preparation: SQL workflows support denormalization, pre-aggregation, cleaning, feature generation, and other transformations used to prepare data for analytics and machine learning.
External data access: SQreamDB can analyze data from multiple sources and formats without requiring all source data to be duplicated. Documentation covers CSV, Avro, Parquet, ORC, and JSON ingestion, as well as Amazon S3 and HDFS environments.
Connectivity: Standard Open Database Connectivity (ODBC), Java Database Connectivity (JDBC), and Python DB-API connections integrate SQreamDB with client applications and analytics tools. Additional connectivity is available for C#, .NET, C++, and Java.
Storage and scaling architecture: Compute and storage are separated, allowing multiple compute units to process data from one or more storage sources. SQreamDB supports horizontal scaling for concurrent statements and uses serializable transactions.
Compression: Ingested data is automatically compressed, with SQream reporting a typical 5:1 compression ratio.

Audience & Use Cases

SQreamDB serves data engineers, data scientists, business analysts, and decision-makers working with large analytical datasets. Common uses include data warehousing, complex SQL analytics, data-lake queries, feature preparation for machine learning, telecommunications analysis, financial analytics, manufacturing data analysis, and retail or advertising workloads.

Deployment and Technical Documentation

SQreamDB is available for private deployments with hardware and installation requirements documented by SQream. The documentation includes setup guidance, SQL execution, client drivers, command-line access, external storage connections, and integration with tools such as Tableau.

SqreamDB

What is SqreamDB?

Categories & Use Cases

Product Demos