Compute Rental
Ctrl K

Guide

Quick Start

GuideIntegrationsFAQSupport
PricingBlog
Quick StartTop-Up and BillingAccelerating Access to Academic ResourcesQuick StartIntroductionMaintenance and TroubleshootingNetwork
PlatformJetBrain ProjectorTmpWebCal Scholars Program @2026Introduction to the Public Beta/Suqian Zone AAbout UsCopying Data Between InstancesAnalysis of Server Performance MetricsTidal Computing PowerLoad Balancing
Domestic ChipsUsing Huawei MindIEHuawei Ascend NPUMooreThread GPU
Environment SetupCUDA/cuDNNMinicondaPython3.XInstalling DependenciesOverviewImages
Enterprise FeaturesFlexible DeploymentElastic Deployment Release NotesBest Practices for Elastic DeploymentPerformance Metrics Monitoring
Container InstancesJupyterLabRemote SSH ConnectionSave the imageScaling ConfigurationMulti-machine, multi-GPU parallel processingDaemonOverviewChange the billing methodMigration Example (Same Region)Migration ExamplesRemote DesktopReset the system
Choosing a GPUGPU SelectionPerformance Testing
DataUpload DataDownload DataPublic DataPublic Cloud Storage (Highly Recommended)Compression / DecompressionFile StorageLocal data diskOverview
Best PracticesFileZillaGitGromacsHuggingFaceKataGoLinux BasicsMPIOpenCLPyCharm Remote DevelopmentR (RStudio) InstallationSSH TunnelTensorBoardRemote Development with VSCodeVisdomVulkanXShellOpen PortsWeChat MessagesPerformanceExpose multiple servicesMoney-Saving TipsComputation Precision IssuesSoftware Sources

Quick Start

Quick Start

07/30/202634055 views

All data (including the environment) in WebCal instances is preserved after shutdown; there is no need to reconfigure or re-upload data upon startup. In short, as long as the instance exists, the data remains intact; however, instances will be released after 15 consecutive days of inactivity. For details, see Instance Data Retention Guidelines.

Create an Instance

After registering, go to the console and click “Rent a New Instance” under the “My Instances” menu.

image_20250226144908

On the instance rental page: Select the billing method, region, GPU model, and number of GPUs, then choose an appropriate available host and image (basic images and community images with various deep learning frameworks pre-installed), and finally click "Create."

If you need a larger hard drive to store data, please note the "Scalable" size in the "Hard Drive" column. For the data disk path, please refer to the documentation.

image_20250226145836 image_20250306103333

After creation, wait for the system to boot automatically. The main access points you’ll use going forward are shown in the screenshot.

Please note that billing begins when the instance status shows "Running." If you are not using the instance, please shut it down promptly to stop billing. For detailed billing rules, see Billing Information.

image_20250226153436

Uploading Data

After booting up, locate the "JupyterLab" shortcut on this running instance, click to open it, and find the upload button in the screenshot below to upload your data. For instructions on uploading folders or using more advanced upload methods, please refer to the Data Upload Documentation.

image_20250226153633

Terminal Training

Open a terminal in the JupyterLab window. If you need to develop remotely using other IDEs, please refer to VSCode (Recommended) and PyCharm.

image_20250226153713

Execute your Python commands in the terminal to complete training, etc.

image_20250226153804

Advanced

  1. How to Choose a GPU
  2. Set Up Your Environment
  3. Upload data via FileZilla
  4. Public Cloud Storage (Highly Recommended)
  5. VSCode
  6. PyCharm
  7. Start the daemon
  8. Clone the code on Git
  9. Linux Basics
  10. Frequently Asked Questions
Next articleSave the image
Compute Rental DocsBack to Guide