2.4 Creating Your First Resource
Creating Your First Resource
Loome lets you create workstations, HPC clusters, and storage through a simplified template-based interface. This walkthrough covers the full process.
Resource Types at a Glance
| Resource | Description | Best For |
|---|---|---|
| Workstations | Virtual machines with desktop environments | Interactive analysis, development, visualization |
| HPC Clusters | Scalable compute clusters | Large-scale simulations, batch processing |
| Storage | Azure storage accounts | Data storage, file sharing |
Creating a Workstation
Step 1: Browse Templates
- Log in to the Loome portal
- Click Templates in the main menu
- Select Workstations category
Step 2: Choose a Template
- Review the available configurations and estimated hourly costs
- Click Select on your chosen template
| Template Type | Description | Best For |
|---|---|---|
| Standard | General-purpose VMs | Development, light analysis |
| GPU-enabled | VMs with GPU acceleration | Machine learning, visualization |
| High Memory | VMs with large RAM | Data-intensive analysis |
Step 3: Configure
| Field | Description | Example |
|---|---|---|
| Name | Unique name for your workstation | my-analysis-vm |
| Size | VM size (CPU, RAM) | Standard_D4s_v3 |
| Operating System | OS image | Ubuntu 22.04, Windows Server |
| Storage | Disk size | 128 GB |
| Region | Azure region | Australia East |
Tip
Choose the smallest VM size that meets your needs. You can resize later if you need more resources.
Step 4: Review and Create
- Review your configuration and estimated costs
- Verify you have sufficient budget
- Click Create to provision the workstation
- Wait for status to change from “Creating” to “Running” (a few minutes)
Creating an HPC Cluster
Step 1: Select Template
- Navigate to Templates > Clusters
- Choose a cluster template (Standard CPU or GPU)
- Click Select
Step 2: Configure
| Field | Description | Example |
|---|---|---|
| Name | Cluster name | my-hpc-cluster |
| Head Node Size | Size of the head/login node | Standard_D4s_v3 |
| Compute Node Size | Size of compute nodes | Standard_HB120rs_v3 |
| Min Nodes | Minimum compute nodes | 0 |
| Max Nodes | Maximum compute nodes | 10 |
| Scheduler | Job scheduler | Slurm, PBS |
Info
Setting Min Nodes to 0 allows the cluster to scale down when idle, saving costs. Nodes spin up automatically when jobs are submitted.
Step 3: Attach Storage
- Select existing storage or create new
- Configure mount points
- Set permissions
Step 4: Create
- Review configuration and costs
- Click Create
- Wait for provisioning (may take 10-15 minutes)
Creating Storage
- Go to Templates > Storage
- Select storage type (Blob or File)
- Configure:
| Field | Description | Example |
|---|---|---|
| Name | Storage account name | myprojectstorage |
| Type | Storage type | Blob, File |
| Redundancy | Data redundancy level | LRS, GRS |
| Size | Initial size allocation | 1 TB |
- Click Create — storage will be available within minutes
Naming Conventions
| Resource | Convention | Example |
|---|---|---|
| Workstations | {project}-{purpose}-vm |
genomics-analysis-vm |
| Clusters | {project}-{type}-cluster |
physics-hpc-cluster |
| Storage | {project}storage |
genomicsstorage |
Warning
Resource names must be unique and follow Azure naming rules: lowercase letters, numbers, and hyphens only. No spaces or special characters.
Your resource is running — time to connect: Connecting to Your Resource