Open Source Projects · Sub-chapter 4 · 3 min read

SODA Foundation

Sub-chapter 4 of Open Source Projects · Open data autonomy across edge, core, cloud

SODA Foundation is the quietest project in this chapter - and a useful one to know about precisely because most engineers don't. Storage is unglamorous. Data management across hybrid environments is unglamorous. But every Welzin customer with multi-cloud or on-prem footprint is wrestling with it, and SODA is the neutral place where vendors and end users try to converge on common APIs.

This is the chapter to read if you're interested in foundational infrastructure that touches everyone.

TL;DR

What - A Linux Foundation umbrella project building an open, unified framework for data and storage management across edge, core, and cloud.
License - Apache 2.0 across the GitHub org.
Origin - Started in 2016 as OpenSDS, rebranded SODA Foundation in 2020. "SODA" is a recursive acronym for SODA Open Data Autonomy.
Focus - data mobility, protection, lifecycle, unified storage, cloud-native storage, governance, orchestration - including data for AI/ML workloads.

Why it exists

Enterprise data is fragmented. A single company runs:

Block, file, and object storage on-prem (often from 3+ vendors)
Multiple public clouds with their own object stores, block volumes, snapshots
Kubernetes clusters needing dynamic CSI provisioners
Edge locations needing tiered storage
An ML platform that wants to read from all of them transparently

Every storage vendor wants its own API to be the universal one - and they all are, at the same time. SODA exists as a neutral space where vendors (Huawei, Fujitsu, NTT, Sony, Toyota, and others have historically participated) and end users build a common abstraction layer so storage and data management don't become a vendor-lock-in tax.

The expanded 2020 charter explicitly covers: data mobility, data protection, data lifecycle, unified storage, cloud-native storage, data governance, data orchestration, and "data energy" - including how data feeds AI/ML workloads.

Architecture

The flagship is the SODA Open Data Framework (ODF) - a unified API layer that sits between applications and storage backends.

text

        Apps / Platforms (K8s, OpenStack, VMs, AI/ML)
                          │
                ┌─────────▼──────────┐
                │   SODA ODF API     │   ← one API for many backends
                └─────────┬──────────┘
                          │
         ┌────────────────┼────────────────┐
         ▼                ▼                ▼
     Block storage   File storage     Object storage
   (on-prem + cloud) (on-prem + cloud) (S3, GCS, …)

Sub-projects you'll see:

Terra - unified storage backend
Strato - multi-cloud connector layer
Delfin - infrastructure management / telemetry (the most active sub-project today)
CSI / Kubernetes connectors - for cloud-native storage flows

Mental model: apps speak one SODA API; SODA translates to whichever underlying storage you have, while collecting metrics and enforcing policy for protection and mobility.

Governance and license

Linux Foundation project with Board, Technical Steering Committee, project-level maintainers.
Apache 2.0 across the GitHub org.
Activity caveat - in 2025–2026, activity is steadier on documentation and Delfin than on new releases. Contributors should expect a smaller, calmer community than CNCF projects. That's an opportunity if you want to be a meaningful voice early.

How to get started

The most approachable sub-project is Delfin (storage telemetry):

bash

git clone https://github.com/sodafoundation/delfin
cd delfin
docker compose up
# Delfin UI on http://localhost:8088

Or read the introduction at docs.sodafoundation.io/introduction before installing anything.

How to contribute

GitHub org: github.com/sodafoundation
Contributor docs: docs.sodafoundation.io (see the "Contribute" section)
Good first issue label is used inconsistently - also check help wanted.
Slack: sodafoundation.io/slack
Mailing lists are linked from the overview page.

A pragmatic first move: improve documentation. SODA's docs are sparse in places, which is a friendly entry-point for new contributors. Documentation PRs are almost always welcome and high-impact.

Hands-on Checkpoints

Read the SODA overview and the 2020 charter expansion press release.
Spin up Delfin locally via Docker Compose.
Browse three SODA sub-project repos. Note their last-commit dates and contributor counts. Be honest about which look healthy.
Find one docs gap or typo. Open a PR.