2.1 Introduction to Loome
What is Loome?
Loome is a cloud resource management platform built on Azure that provides researchers with low-friction access to cloud computing. It offers a single portal for managing projects, computing costs, and elastic demand, allowing you to focus on your research rather than cloud infrastructure.
The Problem It Solves
Researchers need powerful compute — HPC clusters, GPU workstations, scalable storage — but Azure’s native interface is complex and cost management is manual. Loome wraps Azure in a simplified portal with built-in budgeting, curated templates, and on-demand scaling.
Key Benefits
| Benefit | Description |
|---|---|
| Single Console | Manage compute, storage, and costs in one place |
| Simple Provisioning | Deploy from curated templates without Azure expertise |
| Automatic Cost Tracking | Costs tracked against budgets in real-time |
| On-Demand Scaling | Turn resources on/off, scale clusters up/down |
| Curated Templates | Resource templates optimized for specific analysis needs |
Who Should Use It?
Use Loome if you:
- Need Azure-based compute resources (workstations, HPC clusters)
- Want simplified cost management and tracking
- Require HPC clusters or powerful workstations
- Prefer a managed experience over direct Azure access
- Need to scale resources up and down frequently
Consider direct Azure access if you:
- Need advanced Azure services not available through Loome
- Require custom network architectures
- Have Azure expertise and need full flexibility
Loome vs Other Platforms
| Feature | Loome | Compute Service | Azure Direct |
|---|---|---|---|
| Cloud Provider | Azure | AWS | Azure |
| Management Level | Managed | Fully Managed | Self-Managed |
| HPC Support | ✅ Strong | Limited | ✅ Full |
| Cost Tracking | ✅ Built-in | ✅ Built-in | Manual |
| Setup Complexity | Low | Low | High |
| Flexibility | Moderate | Limited | Full |
Common Use Cases
| Use Case | Description | Recommended Resource |
|---|---|---|
| HPC / Simulations | Large-scale simulations and batch processing | HPC Clusters |
| Machine Learning | Training models with GPU acceleration | GPU Workstations |
| Data Analysis | Interactive analysis of large datasets | High Memory Workstations |
| Development | Building and testing research software | Standard Workstations |
| Collaboration | Shared data and environments | Storage + Workstations |
How It Works
- Request Access — Contact the RACE team with your project details
- Log In — Access the Loome portal with your credentials
- Browse Templates — Choose from curated resource templates
- Create — Provision your resource (typically 5-15 minutes)
- Connect — Access your workstation or cluster
- Manage — Start, stop, scale, and monitor costs
Ready to get started? Next up: Getting Access