By clicking “Accept All Cookies”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.

Available now

Introducing HEAVY.AI 7.0

Real-Time Machine Learning Capabilities and so much more....

Read our press release

What's New in HEAVY.AI

NEW HeavyML - Predictive modeling in database

No code joins in Immerse

NEW Resource Manager with CPU Parallelism

HeavyRF cell site editor

Leverage Predictive Analytics on Large Datasets

Orchestrate advanced data science and predictive analytics workflows using SQL or in Heavy Immerse with HeavyML, a suite of new machine learning capabilities available directly in-product. Rapidly cluster data to find outliers, predict missing values, and perform interactive feature engineering, all via the power of SQL. Deliver GPU-accelerated model predictions directly into Immerse.

Learn more in our blog: HeavyML: Deeper Insights with the Power of Machine Learning

Integrate Additional Data on the Fly

It is no longer necessary to pre-create joins for display in Heavy Immerse, as they can be done as needed and persisted into dashboards. Save significant memory, especially when joined on cross filtered data, such as in weather animations or asset performance analyses

High Concurrency and Parallelism via new Executor Resource Manager

New concurrency capabilities in our 7.0 release allow multiple concurrent queries to execute concurrently, including multiple CPU queries alongside a GPU query, optimally using all hardware resources available. Now complex ELT workloads can be run alongside interactive Immerse usage without impacting performance, providing even better performance and scalability than ever before.

Visualize and Simulate Complex Sites & Networks with ML Calibration

Specify and test the impact of multi-antenna site deployments on current and future customer experience. Confidently qualify 5G midband and mmWave deployments by fully accounting for seasonal vegetation attenuation. Use state-of-the-art machine learning techniques in-database to calibrate simulations quickly and easily.