CoSky High-performance, low-cost microservice governance platform (Service Discovery and Configuration Service)

中文文档

CoSky is a lightweight, low-cost service registration, service discovery, and configuration service SDK. By using Redis in the existing infrastructure (I believe you have already deployed Redis), it doesn’t need to bring extra to the operation and maintenance deployment. Cost and burden. With the high performance of Redis, CoSky provides ultra-high TPS&QPS (100,000+/s JMH Benchmark). CoSky combines the process cache strategy + Redis PubSub to achieve real-time process cache refresh, with unparalleled QPS performance (70,000,000+/s JMH Benchmark) and real-time consistency between process cache and Redis.

Service Discovery

Configuration

Examples

Service Consumer --RPC--> Service Provider Examples

Installation

Gradle

Kotlin DSL

    val coskyVersion = "lastVersion";
    implementation("me.ahoo.cosky:spring-cloud-starter-cosky-config:${coskyVersion}")
    implementation("me.ahoo.cosky:spring-cloud-starter-cosky-discovery:${coskyVersion}")
    implementation("org.springframework.cloud:spring-cloud-starter-loadbalancer:3.0.3")

Maven

<?xml version="1.0" encoding="UTF-8"?>

<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">

    <modelVersion>4.0.0</modelVersion>
    <artifactId>demo</artifactId>
    <properties>
        <cosky.version>lastVersion</cosky.version>
    </properties>

    <dependencies>
        <dependency>
            <groupId>me.ahoo.cosky</groupId>
            <artifactId>spring-cloud-starter-cosky-config</artifactId>
            <version>${cosky.version}</version>
        </dependency>
        <dependency>
            <groupId>me.ahoo.cosky</groupId>
            <artifactId>spring-cloud-starter-cosky-discovery</artifactId>
            <version>${cosky.version}</version>
        </dependency>
        <dependency>
          <groupId>org.springframework.cloud</groupId>
          <artifactId>spring-cloud-starter-loadbalancer</artifactId>
          <version>3.0.3</version>
        </dependency>
    </dependencies>

</project>

bootstrap.yaml (Spring-Cloud-Config)

spring:
  application:
    name: ${service.name:cosky}
  data:
    redis:
      url: redis://localhost:6379
  cloud:
    cosky:
      namespace: ${cosky.namespace:cosky-{system}}
      config:
        config-id: ${spring.application.name}.yaml
    service-registry:
      auto-registration:
        enabled: ${cosky.auto-registry:true}
logging:
  file:
    name: logs/${spring.application.name}.log

REST-API Server (Optional)

Installation REST-API Server

Option 1：Download the executable file

Download cosky-server

tar cosky-lastVersion.tar

cd cosky-lastVersion
# Working directory: cosky-lastVersion
bin/cosky --server.port=8080 --spring.data.redis.uri=redis://localhost:6379

Option 2：Run On Docker

docker pull ahoowang/cosky:lastVersion
docker run --name cosky -d -p 8080:8080 --link redis -e SPRING_DATA_REDIS_URL=redis://redis:6379  ahoowang/cosky:lastVersion

Option 3：Run On Kubernetes

apiVersion: apps/v1
kind: Deployment
metadata:
  name: cosky
  labels:
    app: cosky
spec:
  replicas: 1
  selector:
    matchLabels:
      app: cosky
  template:
    metadata:
      labels:
        app: cosky
    spec:
      containers:
        - env:
            - name: SPRING_DATA_REDIS_URL
              value: redis://redis-uri:6379
          image: ahoowang/cosky:lastVersion
          name: cosky
          ports:
            - containerPort: 8080
              protocol: TCP
          resources:
            limits:
              cpu: "1"
              memory: 1280Mi
            requests:
              cpu: 250m
              memory: 1024Mi
          volumeMounts:
            - mountPath: /etc/localtime
              name: volume-localtime
      volumes:
        - hostPath:
            path: /etc/localtime
            type: ""
          name: volume-localtime

---
apiVersion: v1
kind: Service
metadata:
  name: cosky
  labels:
    app: cosky
spec:
  selector:
    app: cosky
  ports:
    - name: rest
      port: 80
      protocol: TCP
      targetPort: 8080

Dashboard

http://localhost:8080/dashboard

Service dependent topology

Role-based access control(RBAC)

• cosky: Reserved username, super user, with the highest authority. When the application is launched for the first time, the super user (cosky) password will be initialized and printed on the console. Don't worry if you forget your password, you can configure enforce-init-super-user: true, CoSky will help you reinitialize the password and print it on the console.

---------------- ****** CoSky -  init super user:[cosky] password:[6TrmOux4Oj] ****** ----------------

• admin: Reserved roles, super administrator roles, have all permissions, a user can be bound to multiple roles, and a role can be bound to multiple resource operation permissions.
• Permission control granularity is namespace, read and write operations

Role Permissions

Add Role

User Management

Add User

Audit Log

Namespace

Config

Edit configuration

Rollback configuration

Import configuration from Nacos

Service

Edit Service Instance

REST-API

https://ahoo-cosky.apifox.cn/

JMH-Benchmark

• The development notebook : MacBook Pro (M1)
• All benchmark tests are carried out on the development notebook.
• Deploying Redis on the development notebook.

ConfigService

gradle cosky-config:jmh
# or
java -jar cosky-config/build/libs/cosky-config-lastVersion-jmh.jar -bm thrpt -t 25 -wi 1 -rf json -f 1

Benchmark                                          Mode  Cnt          Score   Error  Units
ConsistencyRedisConfigServiceBenchmark.getConfig  thrpt       256733987.827          ops/s
RedisConfigServiceBenchmark.getConfig             thrpt          241787.679          ops/s
RedisConfigServiceBenchmark.setConfig             thrpt          140461.112          ops/s

ServiceDiscovery

gradle cosky-discovery:jmh
# or
java -jar cosky-discovery/build/libs/cosky-discovery-lastVersion-jmh.jar -bm thrpt -t 25 -wi 1 -rf json -f 1

Benchmark                                                Mode  Cnt          Score   Error  Units
ConsistencyRedisServiceDiscoveryBenchmark.getInstances  thrpt        76621729.048          ops/s
ConsistencyRedisServiceDiscoveryBenchmark.getServices   thrpt       455760632.346          ops/s
RedisServiceDiscoveryBenchmark.getInstances             thrpt          226909.985          ops/s
RedisServiceDiscoveryBenchmark.getServices              thrpt          304979.150          ops/s
RedisServiceRegistryBenchmark.deregister                thrpt          255305.648          ops/s
RedisServiceRegistryBenchmark.register                  thrpt          110664.160          ops/s
RedisServiceRegistryBenchmark.renew                     thrpt          210960.325          ops/s

CoSky-Mirror (Real-time synchronization of service instance change status)

CoSky-Mirror is like a mirror placed between Nacos and CoSky to build a unified service discovery platform.

Comparison with others

	CoSky	Eureka	Consul	CoreDNS	Zookeeper	Nacos	Apollo
CAP	CP+AP	AP	CP	CP	CP	CP+AP	CP+AP
Health Check	Client Beat	Client Beat	TCP/HTTP/gRPC/Cmd	Keep Alive	Keep Alive	TCP/HTTP/Client Beat	Client Beat
Load Balancing Strategy	Weight/Selector	Ribbon	Fabio	RoundRobin	RoundRobin	Weight/metadata/RoundRobin	RoundRobin
Avalanche Protection	N	Y	N	N	N	Y	N
Auto Logoff Instance	Support	Support	Not Support	Not Support	Support	Support	Support
Access Protocol	HTTP/Redis	HTTP	HTTP/DNS	DNS	TCP	HTTP/DNS	HTTP
Listening Support	Support	Support	Support	Not Support	Support	Support	Support
Multi-data Center	Support	Support	Support	Not Support	Not Support	Support	Support
Cross Registry Synchronization	Support	Not Support	Support	Not Support	Not Support	Support	Not Support
SpringCloud Integration	Support	Support	Support	Not Support	Not Support	Support	Support
Dubbo Integration	Support	Not Support	Not Support	Not Support	Support	Support	Support
K8S Integration	Support	Not Support	Support	Support	Not Support	Support	Not Support
Persistence	Redis					MySql	MySql