feat(routing): 引入多执行池智能调度
将 Worker 发现、路由画像、容量与执行传输抽象为平台无关接口,新增 Kubernetes 和静态容量适配器,并以 shadow 模式接入生产配置。 实现网络与容量评分、路由防抖、池队列、同步 Worker 租约、一次性执行令牌,以及提交状态不明时禁止重复分配的安全语义。 新增 0105 兼容迁移、管理接口、指标、OpenAPI 和回归测试。已执行全量 Go 测试、go vet、OpenAPI、迁移安全、Compose 与 Kustomize 验证。
This commit is contained in:
@@ -47,10 +47,16 @@ data:
|
||||
AI_GATEWAY_ASYNC_WORKER_HARD_LIMIT: "48"
|
||||
AI_GATEWAY_ASYNC_WORKER_GLOBAL_HARD_LIMIT: "48"
|
||||
AI_GATEWAY_ASYNC_WORKER_REFRESH_INTERVAL_SECONDS: "5"
|
||||
AI_GATEWAY_ROUTING_MODE: shadow
|
||||
AI_GATEWAY_ROUTE_PROBE_ENABLED: "true"
|
||||
AI_GATEWAY_ROUTE_PROBE_HOT_INTERVAL_SECONDS: "15"
|
||||
AI_GATEWAY_ROUTE_PROBE_COLD_INTERVAL_SECONDS: "60"
|
||||
AI_GATEWAY_ROUTE_PROBE_TIMEOUT_MS: "3000"
|
||||
AI_GATEWAY_WORKER_ENDPOINT_ALLOW_PRIVATE: "true"
|
||||
AI_GATEWAY_CAPACITY_ORCHESTRATOR_ADAPTER: kubernetes
|
||||
AI_GATEWAY_CAPACITY_POOLS: >-
|
||||
[{"id":"pool-cn-east-1","adapterRef":"ningbo","bootstrapReplicas":1,"minReplicas":1,"maxReplicas":1},{"id":"pool-hk-1","adapterRef":"hongkong","bootstrapReplicas":1,"minReplicas":1,"maxReplicas":1}]
|
||||
AI_GATEWAY_WORKER_AUTOSCALING_ENABLED: "false"
|
||||
AI_GATEWAY_WORKER_REPLICAS_NINGBO: "1"
|
||||
AI_GATEWAY_WORKER_MIN_REPLICAS_NINGBO: "1"
|
||||
AI_GATEWAY_WORKER_MAX_REPLICAS_NINGBO: "1"
|
||||
AI_GATEWAY_WORKER_TARGET_OUTSTANDING_PER_REPLICA: "48"
|
||||
AI_GATEWAY_WORKER_SCALE_UP_WINDOW_SECONDS: "20"
|
||||
AI_GATEWAY_WORKER_SCALE_DOWN_STABILIZATION_SECONDS: "600"
|
||||
|
||||
Reference in New Issue
Block a user