UBIX Makes Complexity Simple by Managing the End-End Data Science and AI Lifecycle
Manage and orchestrate data science storage, engines, and libraries.
- Kubernetes based deployment.
- Pre-configured drag and drop deployment of simple and complex environments.
- Advanced monitoring of instances, applications and processes.
- Manage your cloud instances for all your analytics and AI projects.
Automate data shaping to cleanse, transform, and publish data assets.
- Open-Source big fast data architecture.
- Central management of all your enterprise data for analytics and AI.
- External data feeds can be consumed in combination with enterprise data.
- Create data shapes leveraging a no code wrangling environment.
- Solution templates offer a guide to data management focusing the wrangling around a specific goal or outcome.
Author solution types for specific classes of data science challenges.
- Define end-end workflows to solve specific classes of analytic “Solutions”.
- Organize Solutions by leveraging “Activities” within the workflow.
- Customize “Actions” to execute feature engineering, data visualization, model training and selection.
- Authored workflows create data shape templates within DataSpace.
- Manage ModelSpace libraries of Solutions, Activities, Actions and Action Types to accelerate new Solution Development.
Configure specific analytic workflows to solve specific analytic challenges.
- 100% Code-Free Solution Development.
- Visual interaction layer to capture and interpret data to execute Actions.
- Interact with end-to-end authored workflows to solve specific analytic problems.
- Leverage data shapes created within DataSpace.
- Interact with guided workflows to perform actions for data visualization, data exploration, event discovery, feature engineering, model training and selection.
Analyze and explore data visually without writing code.
- AI API to empower both visualization applications and automation within existing enterprise systems.
- Visualize data and model results across the end-to-end analytic workflows.
- Create dashboards and reports.
- Publish to the web for enterprise sharing.
- Leverage model conclusions to drive automation within other enterprise systems or robotic process automation.
Deploy and manage production models.
- Manage all production analytic assets and models.
- Monitor performance of each component of infrastructure and application.
- Automatic workflow to migrate Solutions into Products.
- Able to pause, activate and archive production assets.
THE UBIX PLATFORM
AI-on-Demand FOR BUSINESS USERS
UBIX | WORK SPACES
Collaborate in an intelligent visual workspace that’s code-free for business users, data engineers and data science teams.
Use open source tools like Jupyter, RStudio, and Python with our big fast data architecture and include custom models in our visual work spaces and workflows.
UBIX | ORCHESTRATED APACHE SPARK PLATFORM
The Ubix Platform wraps and orchestrates multiple open source technologies under a comprehensive, and effective Apache Spark (Lambda) architecture. These technologies support an end-to-end data science process and focus machine learning algorithms on the PROCESS of data science itself.
Data Shaping & Engineering – Automated data shaping speeds ingest, cleanse, join, restructure, and transform tasks without
programming. Dynamic profiling eliminates manual data wrangling.
Task & Process Engine – Enables authoring of classes of data science solutions and then subsequent configuration of end-end analytic workflows by business users to solve specific problems through a code-free visual interface.
Policy & Governance – Documents entire analytic process lifecycle for transparency, reproducibility, and auditability.
API – Provides easy to use, intuitive, and expressive set of routines, protocols, and tools for building and integrating other analytic
applications (RESTful APIs, JSON).
Open Source Storage and ML/DL/AI Libraries
UBIX creates an agile, turnkey approach to big, fast data architecture and data science that helps you start quickly and build intelligently
Apache Spark, Cassandra, Kafka, MongoDB, Elasticsearch, HDFS (vendor neutral), Parquet, TensorFlow, Vowpal Wabbit, SparkML, H2O Sparkling Water, Keras, R, Python
UBIX is available on all cloud hosting platforms and may be deployed in your datacenter/cluster. UBIX may be deployed entirely within its own VPC in your account to provide an additional layer of security and isolation.
Amazon Web Services/ Microsoft Azure / Google Cloud Platform / Red Hat Open Shift / On-premise (Docker-based).