
Data Management for Big Data and Analytics
Delivers accelerated storage to HPC infrastructure workloads.
eXtremeDB for High Performance Computing and Big Data is based on the blazingly fast storage engine for real-time systems, delivers scalability, low latency and advanced analytical capabilities for HPC workloads.
Statistical Analysis
Exploring large amounts of data to discover underlying patterns is a cornerstone of modern business practices. Statistics are applied everywhere from machine learning to finance to digital marketing to healthcare and communications. eXtremeDB HPC comes with a library of more than 100 functions for performing statistical analysis on time series data, such as grand, group, grid and window averages to name a few.
Learn more about what makes eXtremeDB ideal for Time Series Data
Learn more about what makes eXtremeDB ideal for Financial Systems
Review eXtremeDB for HPC FAQ
Pipelining
Pipelining refers to a series of data processing steps where the output of one step is the input to the next step. The pipelining of vector‐based statistical functions is key to eXtremeDB’s ability to accelerate performance when working with IoT and capital markets’ time series data. eXtremeDB’s extensive library of math functions are the building blocks that are assembled into pipelines to minimize data transfers and maximally exploit the CPU’s L1/L2/L3 cache. Don’t see the function your analytics require? No problem, write it yourself and use it alongside the built-in functions seamlessly.
Since 2014, eXtremeDB for HPC has set records, then broken our own records. See the stats.
Watch a 2 minute video on Pipelining.
Review the white paper: Pipelining Vector-Based Statistical Functions for In-Memory Analytics
Pipelining
Pipelining is referred to a series of data processing elements where the output of one element is the input of the next one. Pipelining vector‐based statistical functions is a key to accelerate performance when working with time series data, such as market data. eXtremeDB’s extensive library of math functions are building blocks that can be assembled into a pipeline to minimize data transfers and maximally exploit the CPU and L1/L2/L3 cache.
Since 2014, eXtremeDB for HPC has set records, then broken our own records. See the stats.
Watch a 2 minute video on Pipelining.
Review the white paper: Pipelining Vector-Based Statistical Functions for In-Memory Analytics
Columnar and Row-oriented Storage
Combining analytics and online transaction processing (OLTP) queries is a common requirement for today’s high-performance database systems. eXtremeDB supports two storage formats: Vertical (columnar) storage for time-series data, which is ideal for working with tick streams and machine-generated IoT data, while traditional row-oriented storage is de rigueur for OLTP workloads.
Time series data benefits from columnar data layout
Learn about the advantages of an IMDS for analytics
Learn more about eXtremeDB and in-database analytics
Horizontal Partitioning: Sharding for Elastic Scalability
Efficient Big Data processing frequently requires data partitioning. eXtremeDB offers ultra-fast, elastically scalable data management through sharding — the horizontal partitioning of data. Sharding allows distribution of data over multiple physical nodes, or partitions on the same node, maximizing CPU load and exploiting storage media I/O concurrency.
The developer can specify the storage (in-memory or persistent) for each table, which is ideal for handling real-time and historical data within a single database architecture.
Since 2014, eXtremeDB for HPC has set records, then broken its own records. See the stats.
Scalable distributed database contrast and compare. This table summarizes the primary purpose and characteristics of each distributed database option.
The developer can specify the storage (in-memory or persistent) for each table, which is ideal for handling real-time and historical data within a single database architecture.
Since 2014, eXtremeDB for HPC has set records, then broken our own records. See the stats.
Scalable distributed database contrast and compare. This table summarizes primary purpose and characteristics of each distributed database option.
High Availability Options
eXtremeDB offers a comprehensive set of high availability capabilities through master-slave and multi-master (Cluster) database configurations, and the advanced Active Replication Fabric. Depending on the applications’ requirements, these capabilities can be deployed individually or can seamlessly work together ensuring maximum availability of data and promoting query load balancing in complex heterogeneous environments.
Review a chart of all eXtremeDB distributed database options
Learn how eXtremeDB Active Replication Fabric solves connectivity issues for developers.
Learn more about eXtremeDB Cluster distributed database system for real-time applications
Choice of Programming Interface
eXtremeDB provides a set of libraries written in the C language that enables you to access databases from within any C or C++ program, and language bindings for Python, C#, Java, Scala, PHP, Rust and Lua, and a RESTful API for browser-based apps. Whether you prefer SQL or a native (nonSQL) API, eXtremeDB has you covered. Each language binding defines its own interface to access eXtremeDB databases. SQL and nonSQL APIs, and multiple programming languages can be used simultaneously with the same database.
Read the article, Is SQL Fast Enough for Tick Data?
Learn more about programming with eXtremeDB in our online documentation
Learn more about:
- eXtremeSQL
- eXtremeDB Java Native interface
- eXtremeDB Python interface
- The eXtremeDB type-safe API eliminates database corruption.
- eXtremeDB Native C# interface
Choice of Programming Interface
eXtremeDB provides a set of libraries written in the C language that enables you to access databases from within any C or C++ program. Yet you are not limited to using C/C++. Whether you program in Python, C#, Java, Scala, PHP or Lua, and whether you prefer SQL or a native language no-SQL API, eXtremeDB has you covered. Each language binding defines its own interface to access eXtremeDB databases.
Read the article, Is SQL Fast Enough for Tick Data?
Learn more about programming with eXtremeDB in our online documentation
Learn more about:
- eXtremeSQL
- eXtremeDB Java Native interface
- eXtremeDB Python interface
- The eXtremeDB type-safe API eliminates database corruption.
- eXtremeDB Native C# interface
Used by innovative industry leaders in over 30,000,000 deployments world-wide in these markets and others.
Telecom & Netcom
eXtremeDB powers routers, application traffic management, SCADA network fault management and more from industry leaders like F5, ViaSat, Motorola, Sandvine, Hutchinson, TNS and Ribbon that require real-time performance and five 9s high availability.
Internet of Things
eXtremeDB’s Active Replication Fabric™ manages all aspects of IoT data communication: security, low bandwidth, intermittent connections, multi-tier and bi-directional replication, and more.
Financial Systems
eXtremeDB’s unique hybrid row- and columnar-layout (OLTP and time series) coupled with pipelined functions for statistical analysis and scalable distributed database architecture power record-setting STAC-M3 benchmark results.

Outstanding Structured Database

Best Big Data Analytics & Technology Provider
Related Resources
Articles for Professional Developers
“On Time Series Analysis and Big Data. Interview with Andrei Gorine” – ODBMS.org or “A McObject Focus – What’s Changing in the Satellite Industry” – SatMagazine
See a list of articles
White Papers for Professional Developers
We have been testing, improving on, and retesting our software from the beginning in 2001 in order to provide our clients with the best possible data management solutions. Read “Pipelining Vector-Based Statistical Functions for In-Memory Analytics” and more.
Review our research
Webinars for Professional Developers
Watch to on-demand Webinars, hosted by experts, about proven database management system practices. Watch “Embedded Databases: Building In Always On High Availability” and others.
Review our list of Webinars
Review the Benchmark Test Results
eXtremeDB sets speed records year after year. In multiple independently audited benchmark tests, eXtremeDB has broken its own earlier records for the best (lowest) mean response times and for lowest standard deviation of test results using the SQL database programming language.
