ResearchComputing¶

A unified platform for modern research computing

Welcome to ResearchComputing - an integrated ecosystem of tools designed to make cloud-based research computing accessible, efficient, and cost-effective for scientists, researchers, and research organizations.

Overview¶

ResearchComputing provides a comprehensive suite of cloud-native tools that work together to support the entire research computing lifecycle - from account provisioning and interactive workstations to high-performance batch computing and long-term data archiving.

graph TB
    University[Research Institution] -->|Request Account| Petri[🧫 petri.io]
    Petri -->|Provision| AWS[AWS Account]

    AWS --> CW[🖥️ cloudworkstation.io]
    AWS --> Lens[🔬 lenslab.io]
    AWS --> ATOM[⚛️ atomhpc.io]

    CW -->|Results| Cargo[🚢 cargoship]
    Lens -->|Results| Cargo
    ATOM -->|Results| Cargo

    K8s[On-Prem Kubernetes] -->|Burst| ORCA[🐋 orcapod.io]
    ORCA -->|Scale to| AWS

    style Petri fill:#e1f5ff
    style CW fill:#fff4e1
    style Lens fill:#ffe1f5
    style ATOM fill:#e1ffe1
    style Cargo fill:#f5e1ff
    style ORCA fill:#e1e1ff

The Ecosystem¶

🧫 petri¶

Account & Budget Management

Research-focused AWS account management that simplifies cloud access for academic institutions. Handles account provisioning, budget tracking, and organizational policies.

Use Cases: - University research computing offices - Multi-lab cloud account management - Budget allocation and tracking - Compliance and governance

Learn more about petri →

🖥️ cloudworkstation¶

Interactive Research Workstations

Pre-configured cloud workstations for data science, machine learning, and computational research. Launch GPU-powered environments in minutes with common tools pre-installed.

Use Cases: - Interactive data analysis - Machine learning development - GPU-accelerated computing - Collaborative research sessions

Learn more about cloudworkstation →

🔬 lens¶

Lab Environment Notebook System

Development environments for computational research including Jupyter, RStudio, and VSCode. Designed for researchers who need flexible, reproducible computational environments.

Use Cases: - Jupyter notebook workflows - R statistical analysis - VSCode remote development - Teaching and workshops

Learn more about lens →

⚛️ atom¶

Automated Toolkit for Optimized Modeling

Cloud-native high-performance computing platform for running scientific applications (GEOS-Chem, Gaussian, WRF, VASP) with architecture-optimized containers and cost-efficient job scheduling.

Use Cases: - Atmospheric modeling (GEOS-Chem, WRF) - Quantum chemistry (Gaussian, ORCA) - Materials science (VASP) - Large-scale batch computing

Learn more about atom →

🚢 cargoship¶

Enterprise Data Archiving

S3-optimized long-term data storage and archiving system for research data. Manages data lifecycle, compression, and retrieval for cost-effective long-term storage.

Use Cases: - Research data archiving - Compliance and retention - Cold storage management - Data lifecycle automation

Learn more about cargoship →

🐋 orca¶

Orchestration for Research Cloud Access

Kubernetes-to-AWS burst computing that extends on-premises clusters to the cloud. Seamlessly scale workloads from local infrastructure to AWS when needed.

Use Cases: - Hybrid cloud computing - Kubernetes burst scaling - On-prem + cloud workflows - Cost-effective scale-out

Learn more about orca →

Integration Flow¶

The ResearchComputing ecosystem is designed with integration in mind. Here's how the tools work together:

1. Account Provisioning¶

Start with petri to provision and manage AWS accounts for research groups.

2. Development & Exploration¶

Use cloudworkstation for interactive sessions or lens for notebook-based development.

3. Production Computing¶

Scale to atom for HPC batch workloads, or use orca to burst from on-premises Kubernetes.

4. Long-Term Storage¶

Archive results and datasets with cargoship for cost-effective long-term retention.

Design Principles¶

All ResearchComputing tools share common design principles:

• Cloud-Native: Built for the cloud, not adapted from legacy systems
• Cost-Transparent: Always show estimated and actual costs
• Security First: Private networking, encryption, least-privilege access
• Research-Focused: Designed for scientific workflows and use cases
• Container-Based: Reproducible environments via containers
• Simple UX: Complex infrastructure, simple user experience

Who Uses ResearchComputing?¶

Universities and Research Institutions¶

Provide cloud computing resources to multiple labs and research groups with centralized management and cost allocation.

Research Labs¶

Enable researchers to focus on science rather than infrastructure management, with easy access to interactive and batch computing.

Individual Researchers¶

Access powerful cloud resources without deep AWS knowledge, with tools designed for scientific workflows.

HPC Centers¶

Extend on-premises infrastructure to the cloud for burst capacity and specialized workloads.

Getting Started¶

1. New to cloud research computing? Start with cloudworkstation for interactive sessions
2. Running HPC workloads? Check out atom for batch computing
3. Managing institutional accounts? Explore petri for account management
4. Need development environments? Try lens for Jupyter and RStudio

Community¶

• Blog: researchcomputing.blog - Latest updates and articles
• GitHub: @scttfrdmn - Open source projects
• Twitter: @scttfrdmn - Announcements and updates

About¶

ResearchComputing is developed by Scott Friedman to make cloud-based research computing accessible to scientists and researchers worldwide.

Each tool in the ecosystem is open source and can be used independently or as part of the integrated platform.

Questions? Check out the ecosystem overview or visit individual project pages to learn more.