Famous OSS Web Applications Built with Java and Spring Boot

Spring Boot is still the default way to write Java web applications in 2026. The list of open-source projects built on it is unusually wide: enterprise business apps, CMS, e-commerce, IoT, core banking, identity, and more. There are plenty of reasons to read these repositories: studying architecture, modeling a similar internal system, or learning from production-grade operational patterns.

That said, “famous OSS” covers very different things depending on the goal. Reading PetClinic to learn Spring is not the same as adopting ThingsBoard as an IoT platform, which is not the same as standing up Spring Initializr as an internal developer service. There is also a temporal trap: well-known projects like Eclipse Hono used to be Spring Boot-based but have since moved to Quarkus, which means older write-ups can mislead.

This post lays out famous OSS web applications built with Java and Spring Boot, grouped by category, with notes on architecture and adoption fit.

Official Reference — Spring PetClinic

Spring PetClinic is the canonical Spring Boot sample and the project most newcomers meet first. It models a veterinary clinic and walks through the basics (CRUD, JPA, server-side rendering, and tests). Created in 2003, predating the first Spring Framework release of 2004, it has been rehosted and refreshed alongside successive Spring generations to stay current.

The canonical version is a monolith built on Spring Data JPA and Thymeleaf, but the community maintains forks under the spring-petclinic organization that implement the same domain in microservices, Kotlin, reactive style, and AI-augmented variants built with Spring AI. Being able to read the same business domain implemented in different architectures is what makes it such a strong teaching tool.

It is the first project to point new hires at, and a useful reference when comparing architectural options. Just remember that the production surface area is intentionally trimmed; it is not something you adopt as-is.

Microservices Primer — PiggyMetrics

PiggyMetrics is the other reference worth studying. The domain is personal finance, and the stack is a microservices reference built on Spring Cloud, Eureka, Hystrix, Zuul, MongoDB, RabbitMQ, and Docker Compose. It packs in @RefreshScope for dynamic config, Feign for service-to-service calls, and Sleuth + Zipkin for distributed tracing.

PiggyMetrics is essentially frozen (its last release was Spring Boot 2.0.3 in 2018), so it reads as a study of the Netflix OSS era of Spring Cloud rather than a current reference. PetClinic Microservices uses a more modern stack with Spring Cloud Gateway, Resilience4j, and OpenTelemetry, so reading both is essentially reading the generational shift in Spring Cloud microservices.

CMS — Halo

In the CMS space, Halo has the most momentum right now. The current stable line (Halo 2.x) is built on Java 21+ and Spring Boot 3.x, uses Spring WebFlux and R2DBC for a reactive stack, and ships a Vue 3 frontend with a plugin architecture (the main branch is in the middle of moving to Spring Boot 4.x). GitHub stars sit around 38k, exceptional for a CMS originating from the Chinese OSS community.

Halo’s strategic value is that it broke the “heavyweight Java CMS” stereotype. A single tool covers personal blogs, knowledge bases, corporate sites, and even online stores, and it deploys easily on-prem so you keep data sovereignty. The community edition is fully free OSS and ships with a marketplace for themes and plugins, while the commercial editions add AI-driven site building and regional payment integrations such as WeChat Pay and Alipay.

It is also a strong reference for fully reactive Spring Boot, with WebFlux and R2DBC end-to-end. If you are looking for non-blocking I/O sample code on a real workload, this is one of the cleanest reads.

E-commerce — macrozheng/mall

With around 83k GitHub stars, macrozheng/mall is one of the most prominent Java OSS projects, period. It is a full e-commerce platform built on Spring Boot and MyBatis, with a storefront, an admin panel, and a mobile client built on uni-app. The realistic stack underneath includes Elasticsearch, RabbitMQ, Redis, MongoDB, MySQL, and Docker.

The notable companion is mall-swarm, the microservices variant built on Spring Boot 3.2, Spring Cloud Alibaba (2023.x base), Sa-Token, and Kubernetes. Reading the monolith and the microservices version side by side surfaces the architectural impact of moving the same e-commerce domain from one shape to the other.

It is widely used as a teaching reference, particularly in the Chinese Java community. The caveat is that the documentation is primarily in Chinese, so adopting it in a non-Chinese-speaking team adds a community-language cost to factor in.

For more enterprise-leaning alternatives, Broadleaf Commerce (modular composable commerce, source-available rather than pure OSS) and Shopizer (headless and lean on memory) are the names that come up. Be aware that Shopizer’s public repository sits on the legacy 3.2.x line (the last tagged release is from 2022) while the official site advertises a next-generation Spring Boot 3.x microservices version. Decide which codebase you are actually adopting before you start.

IoT Platform — ThingsBoard

ThingsBoard is the standout industrial IoT platform. It carries about 21k GitHub stars and integrates device management, telemetry ingestion, visualization, and a rule engine into a platform with production deployments handling millions of devices.

The backend is mostly Spring Boot, with some Node.js mixed in. It ships in both monolith and microservices modes; the latter assembles Apache Kafka, Apache Cassandra, Apache ZooKeeper, HAProxy, and Redis or Valkey into a distributed topology. The frontend is Angular.

The Rule Engine is the core of the value proposition. It uses a flow-based programming model that lets operators transform telemetry from MQTT or CoAP in real time, expressed in TBEL (ThingsBoard Expression Language). Rules can be authored and updated without redeploying code, which is a serious productivity advantage on shop floors. The transport layer is Netty-based and supports industrial protocols like Modbus, OPC UA, and BACnet through a gateway.

Adopters cited in the official case studies include Schwarz Group, Medline Adana Hospital, and Super Bock Group. Commit activity is strong. The flip side: running Kafka, Cassandra, multiple services, and a mixed Java / Node.js stack is not a small operational lift, and production adoption needs a real platform team behind it.

Core Banking — Apache Fineract

In financial services, Apache Fineract is treated as a “digital public good.” It is the core banking platform donated by the Mifos Initiative, now governed by the Apache Software Foundation and adopted by financial institutions worldwide.

Technically it requires Java 21+, runs on Spring Boot, and standardizes on PostgreSQL 18+ (the older MySQL / MariaDB path is deprecated). The heart of its scalability is the Spring Batch-driven Close of Business Day (COB) job: interest calculation, automatic withdrawals, and similar daily processing across millions of accounts, distributed across worker nodes, which is heavy by typical web-app standards.

Connection-pool tuning via HikariCP, schema migrations through Liquibase, and a headless API-first design make it a textbook example of robust financial software. The flip side is that double-entry accounting, product modeling, and regulatory requirements come along for the ride. Treat it as a domain adoption rather than a technology adoption.

Identity — Apereo CAS

Apereo CAS is the canonical OSS choice for SSO and identity providers. It supports SAML 2.0, OIDC, OAuth 2.0, MFA, and Spring Webflow, and is widely deployed across enterprises and higher education.

It is built on Spring Boot and Spring Cloud, supports Tomcat / Jetty / Undertow / Docker, and lets you swap authentication backends and data stores per use case. Around 11k GitHub stars, with a near-monthly release cadence. That release density is part of why it is trusted in production.

If your requirements are “stand up an OSS-based company-wide SSO,” “build a higher-ed IdP,” or “consolidate MFA / step-up policies,” CAS is the first candidate. The adoption barrier is requirement design and policy modeling rather than the technology itself, so plan for an investment in operations and policy work, not just deployment.

Healthcare API — HAPI FHIR JPA Server Starter

For FHIR, the standard for healthcare data exchange, HAPI FHIR JPA Server Starter gets you a working server fastest. It bundles HAPI FHIR JPA on top of Spring Boot, runs on H2 by default or PostgreSQL, and ships with Jetty / Tomcat and Docker, a configuration that drops cleanly into research environments and limited production.

The official public demo at hapi.fhir.org is built on these very components, and commercial support is available through Smile CDR, which has accumulated real adoption in healthcare.

One important caveat: the README explicitly says no security implementation is provided. Authentication, authorization, audit logs, and shared caches across multiple instances are all on you. It is the shortest path to a FHIR server PoC, but moving to production means building the safety perimeter yourself.

Microservice Infrastructure — Nacos / Apollo / Sentinel / Seata

The Spring Cloud ecosystem is backed by a cluster of server-side OSS projects, many of which are themselves Spring Boot apps:

Alibaba Nacos: service discovery and configuration. The server is Spring Boot-based and is the heart of Spring Cloud Alibaba. Around 33k stars.
Apollo Configuration Center: distributed configuration management originated at Ctrip. Built on Spring Boot + Spring Cloud, depends only on MySQL, and ships gray-release rollouts, audit logs, and access control out of the box. Around 30k stars.
Alibaba Sentinel: flow control, circuit breaking, and load shedding. Sentinel Core itself is lightweight Java; the dashboard is the Spring Boot application, and it integrates with Nacos / Apollo as data sources.
Apache Seata (Incubating): distributed transactions. The server (TC) is Spring Boot-based and supports AT, TCC, SAGA, and XA modes. Clients integrate via the seata-spring-boot-starter. The project is currently in the Apache Incubator.

These are typically referenced together when you are assembling a Java-native microservice infrastructure or adopting Spring Cloud Alibaba. They have accumulated production hours at Chinese hyperscalers, especially Alibaba’s own use during Singles’ Day (11.11), which is part of why operational confidence is high.

Job Schedulers and Workflows — XXL-JOB / DolphinScheduler

Spring Boot is also strong in batch and workflow territory.

XXL-JOB: a lightweight distributed job scheduler. The combination of xxl-job-admin (Spring Boot management UI) and xxl-job-core (executor library embedded in each worker application) communicates over HTTP RPC. Around 30k stars. In Chinese Java back-office systems, it is effectively the standard scheduler.
Apache DolphinScheduler: a modern DAG-based data orchestration platform. Spring Boot-based api-server / master-server / worker-server, coordinated through ZooKeeper, with Standalone / Cluster / Docker / Kubernetes deployment paths. It is the primary Java / Spring alternative to Apache Airflow.

If you want a DAG-driven Java data pipeline, pick DolphinScheduler. If you want to consolidate cron-style jobs, pick XXL-JOB. The split is roughly that clean.

Developer Platforms — JHipster / Spring Initializr / Spring Boot Admin

The developer-experience side of the ecosystem is also OSS.

JHipster: a generator that produces a complete Spring Boot + frontend (Angular / React / Vue) application skeleton. Spring Security, OAuth2 Resource Server, JPA, Liquibase, Docker, Jest / Cypress / Gatling are all part of the initial scaffold. It is the fastest way to bootstrap an internal admin / CRUD app.
Spring Initializr: the Spring project generation API behind start.spring.io. Self-hosting it inside a company gives you a single point to govern dependency metadata and BOMs as an internal starter distribution platform.
Spring Boot Admin: a multi-app management UI built on Spring Boot Actuator. Heap usage, GC stats, dynamic log-level changes, thread dumps, and similar operational data are all browsable from one place. The 4.x line, with Spring Boot 4 support, shipped in February 2026.
JHipster Online: a web service that wraps JHipster’s generation in a UI. Companies host it internally as a code-generation portal and template-distribution backbone.

The developer-platform group is best understood as products targeted at internal developers rather than business users. Reach for them when you want OSS-grade scaffolding, ops dashboards, and standardization tooling.

API Gateway and Observability — ShenYu / HertzBeat / Pinpoint

A few more Spring Boot-based infrastructure OSS projects are worth flagging: Apache ShenYu for API gateways (Spring Boot 3.x + WebFlux + Project Reactor + Netty, plugin-based architecture), Apache HertzBeat for agentless observability (Spring Boot + JPA + Netty, Prometheus-compatible), and Pinpoint, the APM from Korea’s NAVER (Java agent plus Spring Boot-based Collector / Web on top of HBase).

These are less “Spring Boot web applications” and more “infrastructure built on Spring Boot,” but they have web UIs, are written in Spring Boot, and fit the scope of this post.

Caveat — Famous Java OSS That Are No Longer Spring Boot

To avoid misleading readers based on older write-ups, it is worth being explicit about projects that have moved away from Spring Boot.

Eclipse Hono: 2.0.0 (July 2022) removed the Spring Boot-based AMQP / HTTP / MQTT protocol adapters and Command Router, completing the migration to Quarkus. Container images now ship as Quarkus / GraalVM native.
Apache SkyWalking: a famous Java APM, but the core is built on its own framework, not on Spring Boot. Spring Boot users meet it through agent and integration rather than as a Spring Boot product.

The “older blog post said Spring Boot, but the latest repo is on Quarkus” mismatch (as with Eclipse Hono) is easy to run into. When in doubt, check the current README and pom.xml / build.gradle.

Category Summary

The projects covered above, summarized:

Category	Notable Examples	What They Offer
Official references	Spring PetClinic, PiggyMetrics	Teaching material, internal study groups, architectural baselines
CMS	Halo	Reference for fully reactive Spring Boot 3.x + WebFlux + R2DBC
E-commerce	macrozheng/mall, Broadleaf Commerce, Shopizer	The monolith, composable, and headless flavors
IoT platform	ThingsBoard	Production-grade IoT on Kafka + Cassandra + Rule Engine
Core banking	Apache Fineract	Spring Batch-driven COB jobs and the financial domain
Identity	Apereo CAS	The default OSS SSO with SAML2 / OIDC / MFA
Healthcare API	HAPI FHIR JPA Server Starter	Fastest path to a FHIR server (security is on you)
Microservice infra	Nacos, Apollo, Sentinel, Seata	The server side of the Spring Cloud ecosystem
Jobs / workflows	XXL-JOB, DolphinScheduler	Cron consolidation and DAG orchestration
Developer platforms	JHipster, Spring Initializr, Spring Boot Admin	Scaffolding and ops dashboards as internal products
API gateway / observability	Apache ShenYu, HertzBeat, Pinpoint	Spring Boot-built infrastructure OSS

How to Read This List

With this many candidates, the question is “where do I start.” Picking by goal works well:

Learn Spring Boot architecture: start with Spring PetClinic for the monolith, then expand to PetClinic Microservices and PiggyMetrics for the microservices side.
Bootstrap an internal business app fast: use the JHipster Sample App as the base and trim what you do not need.
Adopt an OSS production platform: pick by domain (ThingsBoard for IoT, Apereo CAS for SSO, Apache Fineract for financial APIs, HAPI FHIR JPA Server Starter for healthcare).
Assemble a microservice infrastructure in OSS: combine the Spring Cloud Alibaba family (Nacos, Sentinel, Seata) with XXL-JOB.
Build an internal developer portal: model it on a self-hosted Spring Initializr plus JHipster Online.

The key call is to decide up front whether you want a reference implementation, a production base, or an internal developer service. PetClinic is a reference, ThingsBoard and Apereo CAS are production bases, and Spring Initializr and JHipster Online are internal developer services. Adopting a heavy domain project (Fineract, ThingsBoard) without crisp requirements tends to fail on the weight of the domain rather than on the technology.

Closing Notes

The set of famous OSS web applications built on Java and Spring Boot stretches from teaching apps to production platforms to developer tools. As of 2026, the practical answer is: PetClinic for learning, ThingsBoard / Apereo CAS / Apache Fineract for production adoption, and JHipster + Spring Initializr for internal infrastructure.

At the same time, projects like Eclipse Hono have moved off Spring Boot, so reading the current README and build files before trusting an older write-up is essential. With Spring Boot 4.0 and Java 25 LTS, the platform has just rolled into its next phase, and new OSS will keep stacking up in this space. It is a good area to revisit periodically, for both hands-on learning and architectural reference.

That’s all from someone surveying Java and Spring Boot OSS examples. From the gemba.