@daocloud-proto/hydra 0.18.0-dev-37 → 0.18.0-dev-41

package/admin-api/maas/v1alpha1/maas.pb.ts

@@ -38,6 +38,8 @@ export enum LoadBalanceStrategy {
   LOAD_BALANCE_STRATEGY_ROUND_ROBIN = "LOAD_BALANCE_STRATEGY_ROUND_ROBIN",
   LOAD_BALANCE_STRATEGY_RANDOM = "LOAD_BALANCE_STRATEGY_RANDOM",
   LOAD_BALANCE_STRATEGY_WEIGHTED = "LOAD_BALANCE_STRATEGY_WEIGHTED",
+  LOAD_BALANCE_STRATEGY_HASH = "LOAD_BALANCE_STRATEGY_HASH",
+  LOAD_BALANCE_STRATEGY_WEIGHTED_HASH = "LOAD_BALANCE_STRATEGY_WEIGHTED_HASH",
 }
 
 export enum RateLimitBaseOn {
@@ -142,6 +144,7 @@ type BaseCreateMAASModelRequest = {
   tokenWeight?: string
   workspaceVisibility?: MAASWorkspaceVisibility
   capabilityRoutes?: MAASCapabilityRoute[]
+  fallbackModel?: UpstreamModel
   modelRatelimitConfig?: ModelRatelimitConfig
 }
 
@@ -173,6 +176,7 @@ type BaseMAASModel = {
   capabilityRoutes?: MAASCapabilityRoute[]
   healthStatus?: MAASModelHealthStatus
   clusterStatus?: HydraManagement_apiWorkspaceV1alpha1Workspace.ClusterStatus
+  fallbackModel?: UpstreamModel
   modelRatelimitConfig?: ModelRatelimitConfig
 }
 
@@ -182,7 +186,6 @@ export type MAASModel = BaseMAASModel
 
 export type LoadBalanceConfig = {
   strategy?: LoadBalanceStrategy
-  weightedConfig?: WeightedLoadBalanceConfig
 }
 
 export type WeightedLoadBalanceConfig = {
@@ -232,6 +235,7 @@ type BaseUpstreamModel = {
   status?: UpstreamModelStatus
   healthCheck?: HealthCheck
   healthMessage?: string
+  weight?: number
 }
 
 export type UpstreamModel = BaseUpstreamModel
@@ -263,6 +267,7 @@ type BaseUpdateMAASModelRequest = {
   tokenWeight?: string
   workspaceVisibility?: MAASWorkspaceVisibility
   capabilityRoutes?: MAASCapabilityRoute[]
+  fallbackModel?: UpstreamModel
   modelRatelimitConfig?: ModelRatelimitConfig
 }
 
@@ -289,6 +294,7 @@ export type ModelRateLimitPolicy = {
   creationTimestamp?: GoogleProtobufTimestamp.Timestamp
   updateTimestamp?: GoogleProtobufTimestamp.Timestamp
   generation?: string
+  description?: string
 }
 
 export type RateLimitValue = {
@@ -372,6 +378,7 @@ export type ModelRateLimitPolicyListItem = {
   ruleCounts?: ModelRateLimitPolicyRuleCount[]
   enabled?: boolean
   updateTimestamp?: GoogleProtobufTimestamp.Timestamp
+  description?: string
 }
 
 export type CreateModelRateLimitPolicyRequest = {
@@ -381,6 +388,7 @@ export type CreateModelRateLimitPolicyRequest = {
   paths?: string[]
   matchStrategy?: RateLimitMatchStrategy
   rules?: ModelRateLimitPolicyRule[]
+  description?: string
 }
 
 export type GetModelRateLimitPolicyRequest = {
@@ -403,6 +411,7 @@ export type UpdateModelRateLimitPolicyRequest = {
   paths?: string[]
   matchStrategy?: RateLimitMatchStrategy
   rules?: ModelRateLimitPolicyRule[]
+  description?: string
 }
 
 export type UpdateModelRateLimitPolicyStatusRequest = {

package/management-api/model_serving/v1alpha1/model_serving.pb.ts

@@ -355,6 +355,15 @@ export type UpdateWSModelServingReplicasRequest = {
   replicas?: number
 }
 
+export type UpdateWSModelServingRuntimeRequest = {
+  workspace?: number
+  cluster?: string
+  namespace?: string
+  id?: string
+  runtimeConfig?: RuntimeConfig
+  resourceConfig?: ResourceConfig
+}
+
 export type ModelServingActionRequest = {
   id?: string
   action?: ModelServingActionRequestAction
@@ -456,6 +465,9 @@ export class WSModelServingManagement {
   static UpdateWSModelServingReplicas(req: UpdateWSModelServingReplicasRequest, initReq?: fm.InitReq): Promise<WSModelServing> {
     return fm.fetchReq<UpdateWSModelServingReplicasRequest, WSModelServing>(`/apis/hydra.io/v1alpha1/workspaces/${req["workspace"]}/model-serving/${req["id"]}/replicas`, {...initReq, method: "PATCH", body: JSON.stringify(req, fm.replacer)})
   }
+  static UpdateWSModelServingRuntime(req: UpdateWSModelServingRuntimeRequest, initReq?: fm.InitReq): Promise<WSModelServing> {
+    return fm.fetchReq<UpdateWSModelServingRuntimeRequest, WSModelServing>(`/apis/hydra.io/v1alpha1/workspaces/${req["workspace"]}/model-serving/${req["id"]}/runtime-config`, {...initReq, method: "PUT", body: JSON.stringify(req, fm.replacer)})
+  }
   static DoModelWSServingAction(req: WSModelServingActionRequest, initReq?: fm.InitReq): Promise<WSModelServing> {
     return fm.fetchReq<WSModelServingActionRequest, WSModelServing>(`/apis/hydra.io/v1alpha1/workspaces/${req["workspace"]}/model-serving/${req["id"]}/actions`, {...initReq, method: "POST", body: JSON.stringify(req, fm.replacer)})
   }

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@daocloud-proto/hydra",
-  "version": "v0.18.0-dev-37",
+  "version": "v0.18.0-dev-41",
   "description": "",
   "author": "",
   "license": "ISC"