finds.dev← search

// the find

KouShenhai/KCloud-Platform-IoT

★ 662 · Java · Apache-2.0 · updated Jun 2026

KCloud-Platform-IoT(老寇IoT云平台)是一个企业级单体架构和微服务架构的IoT云平台。采用DDD(领域驱动设计)思想,基于Spring Boot 4.1.0、Spring Cloud 2025.1.2、Spring Cloud Alibaba 2025.1.0.0 最新版本开发的云服务多租户IoT平台,家人们,点个star!拜托啦~

KCloud-Platform-IoT is a Chinese-authored enterprise IoT platform built on the very latest Spring Boot 4.1/Spring Cloud 2025 stack, supporting both monolith and microservices deployment modes. It wires together an intimidating number of infrastructure components — MQTT via EMQX, two time-series databases (TDengine + TimescaleDB), Kafka, gRPC, Elasticsearch, Nacos — under a DDD/COLA architecture. The target audience is Java teams building multi-tenant IoT backends who want a pre-integrated starting point rather than assembling these pieces themselves.

Tracks bleeding-edge Spring versions (Boot 4.1, Framework 7.0, Security 7.1) more aggressively than most comparable platforms, so you're not inheriting stale dependency debt from day one. The dual monolith/microservices option is genuinely useful — you can start small and restructure later without rewriting business logic. Infrastructure breadth is real: Modbus support, WASM edge gateway in Go, virtual thread support, and GraalVM native image hints are all present, not just listed in a readme. CI is active with Maven, Node, and Go pipelines all green as of last push.

The operational floor is high — the README recommends 32GB RAM for development and says production needs a cluster, which rules this out for small teams or side projects. IoT-specific logic (device provisioning, protocol adapters beyond MQTT/Modbus, time-series query patterns) is thin in the public tree; most of what's visible is generic platform plumbing dressed up with IoT labels. Documentation and issues are almost entirely in Chinese with no English translation path, which is a hard barrier for non-Chinese teams evaluating adoption. The sheer number of dependencies (two message queues, three time-series databases, two serialization libraries) means any real deployment will spend most of its time on infrastructure wrangling rather than building product.

View on GitHub → Homepage ↗

// want more like this?

We dig through GitHub every week and send a few repos picked for what you actually care about — each with an honest take like this one.

Get finds in your inbox → Search again →