CloudML

• 1: Overview
• 2: How-to guides
• 2.1: Kubernetes Cluster Configuration
• 3: API Reference
• 4: CLI Reference
• 5: Release Note

1 - Overview

Service Overview

CloudML is an integrated platform that supports the entire machine learning process—from data analysis to model development, training, validation, and deployment—in a cloud environment.

Features

• Cloud ML is designed to enable users in various roles such as analysts, machine learning engineers, and developers to collaborate in a single environment and easily design and operate machine learning workflows.
• Cloud ML provides an analysis environment based on Python and R, and users with programming experience can leverage the platform more flexibly and effectively. In particular, using the generative AI–based Copilot feature allows code writing, refactoring, error correction, and function recommendation to be performed easily with natural language input, thereby increasing analytical productivity and accessibility.
• Cloud ML systematically supports each stage, including configuring the analysis environment, model development and serving, analysis automation, and visualization. It enables improvements in both productivity and model quality through repetitive experiments and operational automation.

Service Architecture Diagram

CloudML consists of an analysis environment, machine learning lifecycle management, automated analysis support, visualization, and a generative AI‑based Copilot feature, allowing users to perform the entire machine‑learning process in an integrated manner.

Provided features

CloudML provides the following features.

• Visual Modeling: Provides an intuitive interface that lets you build and deploy machine learning models without coding using a Drag&Drop approach. You can easily manage the entire process from data loading to model evaluation and deployment.
• Code-based Development: In the Jupyter Notebook environment, you can freely write and execute code using Python, R, and others. It provides powerful features for advanced users and researchers.
• Workflow Automation: It efficiently automates complex machine learning workflows such as data preprocessing, model training, evaluation, and deployment.
• Experiment Management: You can train machine learning models with various parameter combinations and systematically manage and compare the results.
• Using Copilot Features: It provides a natural-language-based AI assistant that guides and automates the model development process. It supports various tasks such as code generation, refactoring, error correction, and documentation, enhancing productivity.
• Integrated Platform: All features are integrated within CloudML for convenient use.
• Scalability and Flexibility: Supports scaling computing resources and connecting various data sources as needed.

Constraints

Before using CloudML, be sure to check the following constraints and incorporate them into your service usage plan. Since Cloud ML operates in a Kubernetes-based environment, appropriate cluster resource configuration is required for stable service operation.

• Application Basic Resources: To run the Application, a minimum of 24 vCPU cores and 96 GBi of memory are allocated by default.
• Analysis Task Resources: To perform analysis tasks, additional CPU or GPU resource configuration is required beyond the basic resources above. It should be configured appropriately, taking the workload of the analysis tasks into account.
• Copilot (CPU-based usage): To run Copilot on CPU resources, a minimum of 16 vCPU cores and 10 GiB of memory are required. In this case, the CPU resources available for analysis tasks are reduced accordingly.
• Copilot (GPU-based usage): Copilot can also be configured to use dedicated GPU resources.
• Supported LLM models: Currently, the LLM models that can be applied to Copilot are limited to Llama3.

Provision status by region

CloudML is available in the following environments.

Preliminary Service

This is a list of services that must be pre-configured before creating the service. Please refer to the guide provided for each service for details and prepare in advance.

2 - How-to guides

Create CloudML

Users can create the service by entering the required CloudML information and selecting detailed options through the Samsung Cloud Platform Console.

To create a CloudML, follow these steps.

1. Click the All Services > AI/ML > CloudML menu. Navigate to CloudML’s Service Home page.

2. On the Service Home page, click the Create CloudML button. You will be taken to the CloudML page.

3. On the CloudML Creation page, enter the information required to create the service and select detailed options.

   • In the Version Selection area, select the version of the service.

     Category	Required	Detailed description
     Select version	Required	Select CloudML version

     Table. CloudML service version selection options

   • SCP Kubernetes Engine deployment Select the options needed to create a service in this area.

     Category	Required	Detailed description
     Cluster name	Required	Select Kubernetes Engine cluster

     Table. CloudML Service Cluster Selection Options

   • In the Service Information Input area, select the options required to create the service.

     Category	required or not	Detailed description
     CloudML name	Required	Enter service name
     Explanation	Selection	Enter service description
     Domain name	Required	Enter the domain name to be used for the service Enter 2-63 characters using lowercase English letters, numbers, and special characters
     endpoint	Required	Select the endpoint to use in the service Choose between Private and Public
     Copilot	Selection	Select whether to use Copilot in the service Apply when selected requires agreement to terms in the popup window If the selected cluster is not configured with GPUs dedicated to LLMs, or the allocated LLM resources are insufficient, Copilot cannot be applied
     Resource Information	Required	Display resource information of the selected cluster
     Enter SCR information	Required	Enter SCR information to be used in the service Enter private endpoint, authentication key, secret key

     Table. CloudML service information input items

   • Additional Information Input area, please enter or select the required information.

     Category	Required	Detailed description
     tag	Selection	Add Tag Up to 50 can be added per resource After clicking the Add Tag button, enter or select Key, Value values

     Table. CloudML Additional Information Input Items

4. Summary Check the detailed information and estimated billing amount generated in the panel, and click the Complete button.

   • When creation is complete, check the created resources on the CloudML List page.

Check CloudML detailed information

You can view and edit the full list of resources and detailed information for the CloudML service. CloudML Details page consists of Details, Tags, Activity Log tabs.

To view the detailed information of CloudML, follow these steps.

1. Click the All Services > AI/ML > CloudML menu. Navigate to CloudML’s Service Home page.
2. On the Service Home page, click the resource (CloudML) to view detailed information. You will be taken to the CloudML Details page.
   • CloudML Details page displays CloudML’s status information and detailed information, and consists of Details, Tags, Activity History tabs.

     Category	Detailed description
     Service status	CloudML status Creating: Creating Deployed: Created / operating normally Updating: Updating settings Terminating: Terminating Error: Error occurred
     Connection Guide	Service Access Guide Information on host to register on the user’s PC
     Service termination	Cancel Service button

     Table. CloudML status information and additional features

Detailed Information

CloudML List page lets you view detailed information of the selected resource and modify it if necessary.

tag

On the CloudML List page, you can view the tag information of the selected resource, and add, modify, or delete it.

Job History

On the CloudML list page, you can view the operation history of the selected resource.

Terminate CloudML Service

Users can cancel the CloudML service through the Samsung Cloud Platform Console.

To cancel CloudML, follow the steps below.

1. Click the All Services > AI/ML > CloudML menu. Navigate to CloudML’s Service Home page.
2. Click the Cancel Service button on the Service Home page. A service cancellation alert window appears.
3. Enter the CloudML name to delete in the dialog and click the Confirm button.

2.1 - Kubernetes Cluster Configuration

Configuring a Kubernetes cluster

To apply for the CloudML service, a dedicated cluster for CloudML must be set up. A dedicated cluster means creating a Kubernetes Engine that meets or exceeds the required minimum specifications and configuring several necessary settings. Create a dedicated cluster in advance before applying for the CloudML service.

• For instructions on creating a cluster, see the Cluster Creation guide.
• CloudML exposes an HTTPS endpoint on port 443. When creating a cluster, select Public Endpoint.

Recommended specifications for cluster nodes and storage

Cluster nodes can be added or modified after the cluster is created. The following are the recommended specifications for cluster nodes and storage that should be prepared to install CloudML for five users.

Add a label to a node

Add labels to the nodes directly according to the role-specific recommendations in the cluster node and storage specifications.

• For instructions on adding labels to a node YAML, refer to the Edit Node YAML guide.

To add a label to a cluster node, follow these steps.

1. Click the All Services > Container > Kubernetes Engine menu. Navigate to the Service Home page of Kubernetes Engine.
2. On the Service Home page, click the Node menu. You will be taken to the Node List page.
3. On the Node List page, select the cluster for which you want to view detailed information from the gear button at the top left, then click the Confirm button.
4. Select the node you want to view details for and click it. You will be taken to the Node Details page.
5. On the Node Details page, click the YAML tab. You will be taken to the YAML tab page.
6. On the YAML tab page, click the Edit button. The node edit window opens.
7. In the node edit window, add a label that matches the role and click the Save button.
   • Check the following information and add a label that matches the node specifications.

     Category	Purpose-specific labels
     CPU node	App: node.kubernetes.io/nodetype: ml-app Analytics: node.kubernetes.io/nodetype: ml-analytics
     GPU node	Analysis: node.kubernetes.io/nodetype: ml-analytics-gpu Copilot: node.kubernetes.io/nodetype: ml-gpu

     Table. Kubernetes node label items by purpose

3 - API Reference

4 - CLI Reference

5 - Release Note

CloudML