
Plural Stacks: Get Started With One Command
We are excited to announce the release of Plural Stacks. This addition to our platform will help developers deploy open-source infrastructure in a matter of minutes.
Table of Contents
Did you ever wish that installing open-source infrastructure was as simple as one command?
Well, you are in luck.
Today, we are excited to announce the release of Plural Stacks. This addition to our platform will help developers deploy open-source infrastructure in a matter of minutes.
With Plural Stacks, developers can access curated stacks of open-source software, ensuring their infrastructures are built with strong foundations.
Here is why we are excited about Plural Stacks (and why you should be too):
- It allows data professionals to spend less time creating data infrastructure and more time building data pipelines.
- It allows developers to ensure their infrastructure is secure from the ground up.
- It gives developers all the tools they need to launch their open-source infrastructure onto Kubernetes.
What Inspired us to build out this functionality?
We make product decisions based on feedback from our community members and users. We have noticed over the last year that developers repeatedly require a common set of tools. In general, these tools worked in tandem to overcome similar problems.
Additionally, with the combined experience of our team in data infrastructure and DevOps, we have developed preferences for certain technologies and would like to recommend them to Plural users. This allows us to curate the best OSS infrastructure experiences and reduce the number of decisions on the user’s part.
For example, we’ve seen most data professionals deploy an open-source data stack consisting of Airbyte, Dagster, and Apache Superset.
Previously, to deploy this open-source data stack with Plural you would need to write:
plural bundle install airbyte airbyte-{cloud provider)
plural bundle install dagster dagster-{cloud provider)
plural bundle install superset superset-{cloud provider)
Now, with the release of Plural Stacks, that same open-source data stack can be deployed on a Kubernetes cluster with the following command:
plural stack install data
Introducing Plural Stacks
Currently, we have three Plural stacks that are available for public consumption.
Data Stack
Plural’s Open-Source Data Stack consists of the following data tools.
- Airbyte - For data integration
- Dagster - For data orchestration
- Apache Superset - For data exploration and visualization
We chose airbyte because out of all the open source ETL tools, it has the simplest interface and the most active community. Their connector library has grown to a truly immense scale as well, which is what you’d want to perform whatever ETL task you need.
For orchestration, the choice is between Dagster and Airflow, and we believe dagster is architecturally superior. The way you can containerize individual workspaces and connect them to a single dagster instance fits with modern cloud-native best practices, and also allows teams to have separate python dependencies and deploy schedules. In comparison, a monolithic airflow instance doesn’t permit that at all.
For visualization, Superset and Metabase are the two main options, and we lean towards Superset since it’s an apache project and also has a way to support open id connect natively without purchasing an enterprise license. That said, Metabase is still a great choice for many teams.
Get started with the Data stack here.
DevOps Stack
Plural’s DevOps stacks is made of the following software engineering tools.
- Argo CD - For GitOps continuous delivery
- Sentry - For error tracking and performance monitoring
- Grafana - For monitoring and observability
- Kubecost - For visibility into Kubernetes spend and resource allocation
There’s honestly not a lot of controversy here as to why we chose these four tools, they’re all kind of the only game in town.
Get started with the DevOps stack here.
Security Stack
Plural’s security stack gives you end-to-end security for your Kubernetes cluster.
- Kubescape - For a multi-cloud K8s single pane of glass
- Istio - For a service mesh platform
- Vault - For storing identity-based secrets and encryption management
Of these three tools, Istio is the one solution with the most alternatives (there are service meshes from calico, Nginx, and ambassador.) Istio seems to have the strongest open-source community and is freakishly powerful if you get into the weeds of its features. With that said, I do personally think the other service meshes are defensible choices.
Get started with the Security stack here.
How to get started with Plural Stacks
If you are interested in learning more about Plural Stacks and the use cases it can solve for your organization everything we built is open-source, so feel free to check us out on our GitHub and follow along our documentation to get up and running.
Join us on our Discord channel for questions, discussions, and to meet the rest of the community.
Newsletter
Be the first to know when we drop something new.