(link goes to Jetstream2 site)
Today’s cloud service providers offer significant capabilities to accelerate scientific research. However, these capabilities also introduce increased complexity and challenges for research users seeking to use the cloud productively and meaningfully. To address this complexity, research software engineers (RSEs) and scientists often rely on infrastructure-as-code (IaC) or domain-specific language (DSL) recipes that abstract and simplify the intricacies of cloud infrastructure. Yet, cloud recipes can only take software deployments and cloud infrastructure for scientific purposes so far.
CACAO (Cloud Automation and Continuous Analysis Orchestration) is open-source, cloud-native platform that helps users deploy software and infrastructure to multiple clouds using recipes with a mission to:
- adhere to open science standards and FAIR (Findable, Accessible, Interoperable, and Reusable) data principles;
- broaden access to cloud resources for research and educational use cases;
- facilitate collaboration between RSEs, researchers, and educators;
- promote community-driven sharing of best practices for creating and reusing cloud infrastructure recipes.
By focusing on getting stuff done, CACAO helps transform research and education in a multi-cloud world. CACAO is built and maintained by CyVerse, the NSF research project that created Atmosphere.
Key Features
- CACAO helps adapt the cloud to fit your needs. Under the hood, your cloud resources are created using recipes written for Terraform, Ansible, or Kubernetes. If you don’t know how to create your own recipes, don’t worry – Someone else likely created a recipe for the community.
- Future features in CACAO will include support for GCP and Azure Terraform as well as non-Terraform-based recipes, such as Nextflow.
- CACAO will also allow you to activate your resources when your data or workflow code changes and shutdown your resources after executing your workflow, in a Continuous Analysis way.
Existing Recipes
CACAO already supports a variety of recipes, including:
- Zero-to-Jupyterhub
- Kubernetes on demand
- Docker/Dockerswarm
- Magic Castle (from Digital Research Alliance)
- Virtual machines for workshops and classes
- Text-generation-webui
- Danswer
- And many more