Refactored PriorityLevelConfiguration
.. so that concurrency limitation can be done without queuing (like the existing max-in-flight filter does).
This commit is contained in:
Mike Spreitzer
committed by
MikeSpreitzer
MikeSpreitzer
parent
be1658e08c
commit
f9bc2bcf62
67
api/openapi-spec/swagger.json
generated
67
api/openapi-spec/swagger.json
generated
@@ -13992,6 +13992,46 @@
|
||||
],
|
||||
"type": "object"
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.LimitResponse": {
|
||||
"description": "LimitResponse defines how to handle requests that can not be executed right now.",
|
||||
"properties": {
|
||||
"queuing": {
|
||||
"$ref": "#/definitions/io.k8s.api.flowcontrol.v1alpha1.QueuingConfiguration",
|
||||
"description": "`queuing` holds the configuration parameters for queuing. This field may be non-empty only if `type` is `\"Queue\"`."
|
||||
},
|
||||
"type": {
|
||||
"description": "`type` is \"Queue\" or \"Reject\". \"Queue\" means that requests that can not be executed upon arrival are held in a queue until they can be executed or a queuing limit is reached. \"Reject\" means that requests that can not be executed upon arrival are rejected. Required.",
|
||||
"type": "string"
|
||||
}
|
||||
},
|
||||
"required": [
|
||||
"type"
|
||||
],
|
||||
"type": "object",
|
||||
"x-kubernetes-unions": [
|
||||
{
|
||||
"discriminator": "type",
|
||||
"fields-to-discriminateBy": {
|
||||
"queuing": "Queuing"
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.LimitedPriorityLevelConfiguration": {
|
||||
"description": "LimitedPriorityLevelConfiguration specifies how to handle requests that are subject to limits. It addresses two issues:\n * How are requests for this priority level limited?\n * What should be done with requests that exceed the limit?",
|
||||
"properties": {
|
||||
"assuredConcurrencyShares": {
|
||||
"description": "`assuredConcurrencyShares` (ACS) configures the execution limit, which is a limit on the number of requests of this priority level that may be exeucting at a given time. ACS must be a positive number. The server's concurrency limit (SCL) is divided among the concurrency-controlled priority levels in proportion to their assured concurrency shares. This produces the assured concurrency value (ACV) --- the number of requests that may be executing at a time --- for each such priority level:\n\n ACV(l) = ceil( SCL * ACS(l) / ( sum[priority levels k] ACS(k) ) )\n\nbigger numbers of ACS mean more reserved concurrent requests (at the expense of every other PL). This field has a default value of 30.",
|
||||
"format": "int32",
|
||||
"type": "integer"
|
||||
},
|
||||
"limitResponse": {
|
||||
"$ref": "#/definitions/io.k8s.api.flowcontrol.v1alpha1.LimitResponse",
|
||||
"description": "`limitResponse` indicates what to do with requests that can not be executed right now"
|
||||
}
|
||||
},
|
||||
"type": "object"
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.NonResourcePolicyRule": {
|
||||
"description": "NonResourcePolicyRule is a predicate that matches non-resource requests according to their verb and the target non-resource URL. A NonResourcePolicyRule matches a request if and only if both (a) at least one member of verbs matches the request and (b) at least one member of nonResourceURLs matches the request.",
|
||||
"properties": {
|
||||
@@ -14160,21 +14200,29 @@
|
||||
"type": "object"
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.PriorityLevelConfigurationSpec": {
|
||||
"description": "PriorityLevelConfigurationSpec is specification of a priority level",
|
||||
"description": "PriorityLevelConfigurationSpec specifies the configuration of a priority level.",
|
||||
"properties": {
|
||||
"queuing": {
|
||||
"$ref": "#/definitions/io.k8s.api.flowcontrol.v1alpha1.QueuingConfiguration",
|
||||
"description": "`queuing` holds the configuration parameters that are only meaningful for a priority level that does queuing (i.e., is not exempt). This field must be non-empty if and only if `queuingType` is `\"Queuing\"`."
|
||||
"limited": {
|
||||
"$ref": "#/definitions/io.k8s.api.flowcontrol.v1alpha1.LimitedPriorityLevelConfiguration",
|
||||
"description": "`limited` specifies how requests are handled for a Limited priority level. This field must be non-empty if and only if `type` is `\"Limited\"`."
|
||||
},
|
||||
"type": {
|
||||
"description": "`type` indicates whether this priority level does queuing or is exempt. Valid values are \"Queuing\" and \"Exempt\". \"Exempt\" means that requests of this priority level are not subject to concurrency limits (and thus are never queued) and do not detract from the concurrency available for non-exempt requests. The \"Exempt\" type is useful for apiserver self-requests and system administrator use. Required.",
|
||||
"description": "`type` indicates whether this priority level is subject to limitation on request execution. A value of `\"Exempt\"` means that requests of this priority level are not subject to a limit (and thus are never queued) and do not detract from the capacity made available to other priority levels. A value of `\"Limited\"` means that (a) requests of this priority level _are_ subject to limits and (b) some of the server's limited capacity is made available exclusively to this priority level. Required.",
|
||||
"type": "string"
|
||||
}
|
||||
},
|
||||
"required": [
|
||||
"type"
|
||||
],
|
||||
"type": "object"
|
||||
"type": "object",
|
||||
"x-kubernetes-unions": [
|
||||
{
|
||||
"discriminator": "type",
|
||||
"fields-to-discriminateBy": {
|
||||
"limited": "Limited"
|
||||
}
|
||||
}
|
||||
]
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.PriorityLevelConfigurationStatus": {
|
||||
"description": "PriorityLevelConfigurationStatus represents the current state of a \"request-priority\".",
|
||||
@@ -14194,13 +14242,8 @@
|
||||
"type": "object"
|
||||
},
|
||||
"io.k8s.api.flowcontrol.v1alpha1.QueuingConfiguration": {
|
||||
"description": "QueuingConfiguration holds the configuration parameters that are specific to a priority level that is subject to concurrency controls",
|
||||
"description": "QueuingConfiguration holds the configuration parameters for queuing",
|
||||
"properties": {
|
||||
"assuredConcurrencyShares": {
|
||||
"description": "`assuredConcurrencyShares` (ACS) must be a positive number. The server's concurrency limit (SCL) is divided among the concurrency-controlled priority levels in proportion to their assured concurrency shares. This produces the assured concurrency value (ACV) for each such priority level:\n\n ACV(l) = ceil( SCL * ACS(l) / ( sum[priority levels k] ACS(k) ) )\n\nbigger numbers of ACS mean more reserved concurrent requests (at the expense of every other PL). This field has a default value of 30.",
|
||||
"format": "int32",
|
||||
"type": "integer"
|
||||
},
|
||||
"handSize": {
|
||||
"description": "`handSize` is a small positive number that configures the shuffle sharding of requests into queues. When enqueuing a request at this priority level the request's flow identifier (a string pair) is hashed and the hash value is used to shuffle the list of queues and deal a hand of the size specified here. The request is put into one of the shortest queues in that hand. `handSize` must be no larger than `queues`, and should be significantly smaller (so that a few heavy flows do not saturate most of the queues). See the user-facing documentation for more extensive guidance on setting this field. This field has a default value of 8.",
|
||||
"format": "int32",
|
||||
|
||||
Reference in New Issue
Block a user