TechPrufer
Get started
Hex
Hex logoHex · Feature trail

Advanced Compute Profiles

3 posts · Oct 2023 – Dec 2024

What this trail covers

Hex has evolved its compute capabilities by introducing Advanced Compute Profiles, offering users significantly more processing power (up to 16 CPUs and 128GB RAM, or V100 GPUs) for demanding data science and ML workflows directly within the Hex notebook environment. This complements existing integrations with data warehouses and external ML platforms, allowing for end-to-end model development without data export or environment reconfiguration. New features include adjustable idle timeouts for kernels, pay-per-minute billing, and admin controls for spend limits and access.

All Hex trails
  1. Introducing Advanced Compute ProfilesThis post introduces Advanced Compute Profiles for Team and Enterprise customers, offering up to 16 CPUs/128GB RAM or V100 GPUs. It details the new 'idle timeout' setting for kernels (15 mins to 24 hours), pay-per-minute billing, and admin controls for spend limits and access. It also highlights how this complements existing integrations like Snowpark ML and Modelbit, and allows for prototyping within Hex before pushing to other systems for inference.
  2. Building ML Models In Minutes With Hex and Snowflake ML FunctionsThis post details how Hex integrates with Snowflake ML Functions to enable building and deploying ML models using only SQL. It showcases the ease of use of Snowflake's forecasting function within Hex, including partitioned model training for handling multiple series. The post highlights the ability to train and infer models with minimal SQL code, leveraging Hex's environment for data preparation and Python integration.
  3. Introducing Hex 3.0Introduced parallel execution for the compute engine, enabling simultaneous execution of cells, particularly beneficial for complex projects with expensive queries. This feature is live for warehouse SQL queries and rolling out to other cell types. Also introduced native Python dataframe pushdown, allowing massive data workflows and ML pipelines to run seamlessly on leading cloud data infrastructure solutions.