GreenNode AI Platform Release Note 2024

GreenNode AI Platform Release Note 2024

This central hub provides comprehensive information about the latest updates, new features, enhancements, and bug fixes introduced in each release of the GreenNode AI Platform in 2024. Our goal is to keep you informed and empowered to make the most of the platform's capabilities. We recommend reviewing the release notes for the latest version to stay up-to-date with the platform's evolution and take advantage of the new features and improvements.
  1. Version 4: (Released November 25)
    1. Supervised Tuning: Fine-tune models on supervised learning tasks by adjusting hyperparameters like learning rate, batch size, and optimizer.
    2. ​RLHF Tuning: Optimize models for reinforcement learning from human feedback (RLHF) tasks, focusing on parameters related to reward modeling and policy optimization.
  2. Version 3.0 (Released October 14)
    1. Usage Reports: Generate comprehensive reports for each resource type, including compute instances (GPU, CPU) and storage volumes.
    2. Action-Based Tracking: Monitor resource usage based on specific actions, such as training jobs, inference tasks, and data processing.
  3. Version 2.0 (Released September 11)
    1. Significant updates and enhancements to the platform.
    2. Detailed as table below
  • Version 1.0 (Released July 16)
    • Initial release of the Greennode AI Platform.
    • Core features and functionalities introduced.
    • Detailed as table below
Release Date
Feature
Description
Reference
November, 25
Model tuning
The Model Tuning feature empowers users to fine-tune machine learning models by optimizing hyperparameters. This enables users to tailor models to specific datasets and tasks, leading to improved performance and accuracy.
  1. Flexible Hyperparameter Tuning: Explore a wide range of hyperparameters and their combinations to achieve optimal results.
  2. Efficient Training: Leverage distributed training and hardware acceleration to speed up the tuning process.
  3. Robust Logging: Track the progress of tuning jobs.
  4. User-Friendly Interface: A simple and intuitive interface for creating, managing, and monitoring tuning jobs.
Model tuning guideline: https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/model-tuning/guides
October, 14
Usage report

Usage Report provides a detailed breakdown of your resource consumption, helping you optimize costs and performance. By tracking the usage of compute instances & storage volume, you can identify areas for improvement and make data-driven decisions.

  • Real-time Monitoring: Track resource usage in real-time to ensure optimal resource allocation.
  • Cost Breakdown: Understand the cost breakdown based on different resource types and usage patterns.
Usage report detailed guide: https://helpdesk.greennode.ai/portal/en/kb/articles/usage-report
October, 14
Action history

Action history provides a detailed record of user interactions with resources within the platform. This feature allows you to:

  • Monitor User Behavior: Track the actions performed by users, such as creating, editing, or deleting resources.
  • Identify Usage Patterns: Analyze user behavior to identify trends and optimize resource allocation.
  • Debug Issues: Review the history of actions to troubleshoot problems and identify the root cause of errors.
  • Audit and Compliance: Maintain a comprehensive audit trail for compliance and security purposes.
Action history detailed guide: https://helpdesk.greennode.ai/portal/en/kb/articles/action-history
September, 11
CPU instance

This feature allows to create CPU instance with a base image or Jupyter notebook image and you can use this instance for following main purposes:

  • Mount network volume.

  • Collect data from sources such as downloading from public sources or uploading from your local machine.

  • Preprocess data.

  • Process ML/AI tasks no requiring heavy computational power.

CPU instance types on demand: https://helpdesk.greennode.ai/portal/en/kb/articles/compute-instances
September, 11
SSH GPU / CPU container instance
This feature allows to connect to the notebook instance using Secure Shell (SSH), providing a more secure and flexible connection method.
Connect to a notebook instance: https://helpdesk.greennode.ai/portal/en/kb/articles/connect-to-notebook-instances
September, 11
Fix bugs & improvement
Fix some bugs and improve performance features

September, 11
Network volume

Network Volumes on the Greennode AI Platform provide a high-performance and scalable storage solution specifically designed to meet the storage and data management needs of AI resources. One of the standout features of Network Volumes is their ability to be accessed flexibly and easily from various components within the platform, including:

  • Notebook Instances: Interactive working environments where data scientists and AI engineers can write, run, and debug code.

  • Training Jobs: Model training jobs where AI models are trained on large datasets.

  • Online Prediction: API endpoints that allow deployed models to make real-time predictions.

Guides: https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/network-volume/guide
July, 16
Notebook instance

  • This is an interactive environment where data scientists and engineers can write, execute, and iterate on code. It usually includes features for data exploration, visualization, and collaborative work. Notebooks often support multiple programming languages like Python, R, or Scala and facilitate the development and testing of algorithms and models.

  • Current version is only supporting for the Jupyter Notebooks.

Guides: https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/notebook-instance/guides

Tutorials: https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/notebook-instance/tutorials
July, 16
Model training

  • Model training involves using datasets to create and optimize machine learning models. This process occurs in the cloud environment, where data scientists run code to build models and tune hyperparameters. Training uses computational resources like GPUs or TPUs to accelerate the learning process and improve model accuracy.

  • Current version is only supporting for Training Frameworks: TensorFlow, PyTorch, or Scikit-learn.

July, 16
Distributed model training

Distributed training involves training a machine learning model across multiple GPU instances simultaneously.

Guides: https://helpdesk.greennode.ai/portal/en/kb/articles/create-a-distributed-training-with-pre-built-container
July, 16
Model Registry

The model registry is a centralized repository for storing trained models, their metadata, versions, and associated artifacts. It allows for version control, management, and organization of models developed during the training phase. This enables easy access, retrieval, and deployment of models for various purposes.

Guides: https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/model-registry/guides

Tutorials: https://helpdesk.greennode.ai/portal/en/kb/articles/accessing
July, 16
Inference
Deploy and manage AI models for lightning-fast prediction including simplified model deployment and scalable infrastructure.
July, 16
Integrate billing payment with Stripe

  • Pay for payments using the Pay As You Go method.

  • Pay bills with payment provider Striple.

https://helpdesk.greennode.ai/portal/en/kb/greennode-ai-platform/billing-method
July, 16
IAM

  • Supports IAM hierarchy of GreenNode AI Platform features including Notebook instance, Model training, Model registry and Model serving (Online prediction).

  • Currently, GreenNode IAM Platform supports two types of users: IAM Users & Root Users.


    • Related Articles

    • GreenNode AI Platform Release Notes 2025

      This central hub provides comprehensive information about the latest updates, new features, enhancements, and bug fixes introduced in each release of the GreenNode AI Platform in 2025. Our goal is to keep you informed and empowered to make the most ...

    • Greennode Instances Type

      In essence, tasks requiring heavy computational power, parallel processing, or dealing with large datasets and complex algorithms can benefit from using instances equipped with high-performance GPUs. These instances accelerate computations and ...

    • Create a Network Volume

      How to Create a Network Volume on Greennode AI Platform? This guide will walk you through the process of creating a new volume on your AI platform to store data, models, and other necessary files. Step 1: Access the Volume Creation Interface Log in: ...

    • Create a network volume

      How to Create a Network Volume on Greennode AI Platform? This guide will walk you through the process of creating a new volume on your AI platform to store data, models, and other necessary files. Step 1: Access the Volume Creation Interface Log in: ...

    • Payment Method

      Depositing Credits to Your GreenNode Account via Stripe This guide provides a step-by-step walkthrough of how to deposit credits into your GreenNode account using Stripe. Having a credit balance allows you to access and utilize GreenNode's AI ...