Over the last few months, Bodo and SaturnCloud have worked together to provide a joint solution with Bodo software running within Saturn Cloud resources. Data scientists can now access Bodo for extreme performance Python in Saturn Cloud's collaborative environment and computing resources, all in one place. Users can scale prototypes to petabyte-scale parallel-processing production environments without any tuning or re-coding. Saturn Cloud provides a pre-built template with Bodo already installed and configured. Within this template, users can access the functionality of Bodo within JupyterLab or via SSH from VSCode, PyCharm, or the terminal. Users are able to get up to 4TB of RAM and 128 vCPUs on SaturnCloud to run their Bodo workloads.
We also provide examples to get you started:
- Use Bodo to speed up feature engineering and model training
- use Bodo to speed up data manipulation and analysis.
The Bodo platform democratizes access to parallel computing with the flexibility and simplicity of native Python, while providing extreme performance for data analytics, ML and AI at massive scale. At its core, Bodo offers a JIT compiler, which compiles Python code into automatically parallelized, efficient machine code. With Bodo, developers can build and develop applications on their laptop in a familiar development environment using native Python without any additional API layers. They can then deploy the same code on clusters with thousands of cores without any re-coding or parameter tuning.
Bodo is the solution for data teams looking for a smooth workflow from prototype to production, with a focus on solving the business problem rather than low-level configuration and engineering details. Saturn Cloud is a managed data science and machine learning platform which automates DevOps and ML infrastructure engineering. We found that there is a natural synergy between Bodo and Saturn Cloud, as Saturn Cloud’s pre-built tools enable the developer to use their existing workflow, while providing an environment where they don’t need to rely on dev sources or manage compute environments.