The Kinetica Engine

With a memory-first, GPU database at its core, the Kinetica technology combines advanced analytics, visual discovery, location intelligence, and machine learning within a single engine

In the Extreme Data Economy, data comes from unpredictable sources and analysis becomes much more complex.

Data can be big or small, static or streaming, structured or unstructured, human or machine. On top of that, it’s perishable: data’s useful life may only be hours or minutes, weeks or months or more. The siloed legacy databases built for serial computing that many businesses still have in place just can’t keep up.

CHALLENGES WITH EXISTING TECHNOLOGIES

Poor Performance

Performance can’t keep up with extreme data: structured, unstructured, location, time, static or streaming.

Unwieldy Streaming Data

Batch processing leads to stale data and delayed decision making.

Too Complex

With lots of data technologies strung together, deployments become too complex.

Lacking AI Support

Minimal support for advanced algorithms, machine learning & deep analytics within the database.

The Kinetica Solution

The Kinetica engine intelligently combines GPUs, CPUs, and in-memory computing to analyze complex, unpredictable, streaming data with dramatically faster performance.

OLAP Query Engine

Perform real-time analytics on streaming or static data with up to 100 times the performance of legacy systems

Visual Rendering

Instantly generate and render visuals on billions of location points in milliseconds for faster dashboarding and discovery

Geospatial Analytics

Process location data as a native data type to perform advanced location analytics on massive datasets

Machine Learning Pipeline

Train models and run inference faster on the entire data corpus, not just small subsets with integrated ML workflows

Inside Kinetica
Visualization
and client apps
Insight engine
Deployment options:
on-premises or cloud

Kinetica Reveal

BI Tools

Custom Apps

API Layer

OLAP
Query Engine

OLAP Query Engine

Perform real-time analytics on streaming or static data with up to 100 times the performance of legacy systems.

Visual
Rendering

Visual Rendering

Instantly generate and render visuals on billions of location points in milliseconds for faster analytical dashboarding and discovery.

Geospatial
Analytics

Geospatial Analytics

Process location data as a native data type to perform advanced location analytics on massive datasets.

ML
Pipeline

Machine Learning Pipeline

Train models and run inference faster on the entire data corpus, not just small subsets, with integrated machine learning workflows

In-Memory GPU Database In-Memory GPU Database

Crunch large volumes of data fast with GPUs that work in parallel. Accelerate location-based and in-memory analytics, machine learning, and AI.

Parallel Orchestration Engine Parallel Orchestration Engine

Abstracts the hardware and compute architectures to distribute workloads optimally across GPUs and CPUs.

Key Capabilities

Distributed Scale Out

Kinetica can scale up or scale out to support any dataset size. This leads to smaller and more predictable hardware costs.

Text Search

Full text search functionality on the data itself, supported by Natural Language Processing (NLP).

Native Geospatial Capability

Make location a first class citizen. Run geospatial functions for filtering massive datasets by area, by track, custom shapes, and more exponentially faster than traditional systems.

Visual Dashboards

Allow users to easily and intuitively slice and dice data and visualize datasets with Reveal, a built-in interactive dashboard.

SQL

Run a full range of SQL queries through our ODBC or JDBC connector on static or streaming data to process and analyze billions of rows in milliseconds.

Connectors

Pre-built connectors for Apache Kafka, Apache NiFi, Apache Spark, Streamsets, and ODBC/JDBC make it simple to ingest data from multiple sources.

UDFs and Model Frameworks

Program with an extensible and highly flexible framework to perform advanced analytics and machine learning with data stored in Kinetica, taking full advantage of a distributed architecture.

APIs & SDKs

Build custom operational analytical application using a wide range of Open Source native language bindings including Java, Python, Javascript, and C++ or using the native RESTful API.

Test Drive Kinetica Free for 30 days!

The Kinetica engine is built for the enterprise,
with compression, high availability, and enhanced security

Reference Architecture

Application Layer

Ingestion

Insight Engine

Transactional

Data Lake

Optimize performance with a single insight engine with
a GPU database at its core.

Ingest and analyze streaming data from unpredictable sources in parallel,
in real-time, so you can capitalize on it now.

Analyze Streaming Data

Making it possible to analyze streaming and historical data together for the most accurate and more real time view of the world.

Ingest Data from
Ever-changing Sources

With new data types like location and time and with constantly emerging data sources, you need a single engine that can support it all.

Improve Performance

Kinetica moves faster because we’re an insight engine powered by GPUs, running at the speed of memory with a built-in database, so you can take in and analyze data in parallel, in real-time.

Simplify Complex Analysis

Analyze data aggregated from hundreds of sources in real-time, combining complex algorithms and machine learning within the database without needing to integrate several different data management systems.

So much raw compute power, you won't need to
think about indexing, partitioning or downsampling!

Free Technical Whitepaper

Inside Kinetica

Take a deep dive into the internal architecture of Kinetica