Corporate AI made efficient
Innovations for building the Liquid Big Data Continuum
Contextual Decision Making
Alternative Frames provides unified access to multi-dimensional open data sources beyond the limited boundaries of enterprise data.
Complementary facts, events, trends and other contextual data are made ready for DataFrame based data science and turn enterprise data into understandable and self-explanatory assets.
Structured AI Solution Language (SAISL)
AI is at a crossroads like structured data analytics before. Handcrafted data science is the opposite of being a future-oriented tool@scale for millions of lightweight apps.
SAISL makes the entire AI triad declarative, from data fusion to modeling to inference, ready for automated code generation of prediction apps.
Declarative AI defines the next AI maturity level: "Predict as you train" with reusable AI recipes included.
AI Data Flows Implemented
The AI data flow paradigm shifts traditional algorithm centricity to application centricity with a fast & flexible journey from data to algorithms and solutions.
Lightweight smart data flows are at the heart of this paradigm and represent the ultimate business assets in an agile AI environment.
Deep Pipes implements the AI data flow paradigm on top of Google's Data Fusion platform and thereby integrates with Declarative AI.
More than 200 pre-built configurable flow components, organized as connectors, transformers and operators, transform the creation of AI applications into a no-code plug & play experience.
deep connect
DeepConnect provides 50+ plug & play connectors for breaking purpose built data siloes in batch and realtime processing.
This module implements one of the key building blocks of the new data flow paradigm that shifts todays' AI thinking from algorithm to solution centricity.
deep transform
This module provides 75+ plug & play data extractors & transformers for integrating feature engineering with advanced data intelligence.
DeepTransform complements DeepConnect and prepares batch and realtime data to match the different needs of DataFrame based big data analytics.
deep operate
DeepOperate provides 75+ plug & play for supporting the full spectrum of modern data intelligence.
This module integrates with DeepTransform and brings DataFrame based operators for deep & machine learning, natural language & time series, and also SQL and rule processing.
Unified Graph Analytics
DeepGraph provides the most comprehensive collection of Graph algorithms with unified GraphFrame based access.
DeepGraph seamlessly integrates with various next-generation graph databases like Dgraph, Graphs Ignite and HGraphDB for graph persistence.
Combined with modules like DeepPipes and Works. Ignite, DeepGraph is key to provide unified access along the entire journey from data to information to knowledge.
In-Memory Speed CY-IOT Factory
Cyber Defense can longer be an afterthought when shaping the world of Smart AgriFood, Cities, Energy, Industry and Water.
CY-IOT integrates Endpoint & Network Security, Threat Intelligence and FIWARE compliant Smart Environments with high performance in-memory speed data computing.
Works. Ignite turns Apache Ignite into an in-memory graph database to support the future of connected and cyber resilient devices. NGSI-LD data from AgriFood to Water can be linked with cyber attack signals and threat events for advanced graph traversal and analytics.
fiware ignite
Realtime Ignite streamer that integrates Fiware Brokers with Apache Ignite.
Brokers are key components for data sharing in Fiware enabled smart agrifood, city, energy and industry environments.
opencti ignite
Realtime Ignite streamer that integrates Open Cyber Threat Intelligence with Apache Ignite.
OpenCTI is a unified platform for all levels of threat intelligence and is vital for classifying data like anomalies as malicious or not.
osquery ignite
Realtime Ignite streamer that integrates Osquery based endpoint security with Apache Ignite.
Osquery monitors 10,000+ endpoints to detect suspicious signals of cyber attacks and potential intrusions.
zeek ignite
Realtime Ignite streamer that integrates Zeek based network security with Apache Ignite.
Zeek monitors huge amounts of network traffic data to detect suspicious signals of cyber attacks and potential intrusions.
alpakka ignite
This module provides an Apache Ignite Sink for stream-aware, reactive ingestion pipelines.
It is built with built-in support for backpressure, and prolongates Alpakka's realtime streaming to reach high performance data computing.
graphs ignite
This module complements Apache Ignite with a distributed in-memory Graph database.
Apache Ignite transforms into an in-memory speed data computing platform with support for DataFrames, GrahFrames and Graph Traversals.
superset ignite
This module provides a SQLAlchemy driver for Apache Ignite.
This driver was built for data exploration and visualization of Ignite datasets with Apache Superset.
thingsboard ignite
This module integrates the Open IoT platform ThingsBoard with Apache Ignite.
Device & sensor data from plenty of ThingsBoard platforms are accessible to data exploration, machine learning and graph analytics.
Unified Realtime Processing
Works. Metastream extends the reach of Apache Spark Streaming and Structured Streaming with a diversity of streaming connectors and SQL data destinations.
Metastream supports event streaming with lightweight data shippers for ad-hoc realtime monitoring, seamless integration with Declarative AI & DeepPipes, and SQL based complex event processing.
Structured Streaming extensions bring together realtime streaming, DataFrame based data science and open source & commercial SQL data sinks.
works. beats
This module is part of the CY-IOT stack and provides purpose-built, standalone Akka based micro-services for data acquisition.
Works. Beats consumes sources like industrial OPC-UA servers and exposes them as MQTT and Server Sent Events to build lightweight data monitoring applications.
works. stream
This module integrates with Works. Beats and complements Apache Spark Streaming with purpose-built CY-IOT streaming receivers.
Works. Stream was implemented to publish unified Cy-IOT events as a single streaming source for DeepPipes and make them accessible for full-spectrum data intelligence.
works. sqlstream
This module complements Apache Spark SQL streaming with purpose-built streaming readers and writers.
Works. SqlStream integrates with Works. Stream and supports various open source and commercial data sinks like Aerospike, Exasol, SAP HANA and Snowflake.