REST Resource: projects.locations.reservations

Resource: Reservation

A reservation is a mechanism used to guarantee slots to users.

JSON representation 
{
  "name": string,
  "slotCapacity": string,
  "ignoreIdleSlots": boolean,
  "autoscale": {
    object (Autoscale)
  },
  "concurrency": string,
  "creationTime": string,
  "updateTime": string,
  "edition": enum (Edition),
  "primaryLocation": string,
  "secondaryLocation": string,
  "originalPrimaryLocation": string,
  "scalingMode": enum (ScalingMode),
  "maxSlots": string
}
Fields
name

string

The resource name of the reservation, e.g., projects/*/locations/*/reservations/team1-prod. The reservationId must only contain lower case alphanumeric characters or dashes. It must start with a letter and must not end with a dash. Its maximum length is 64 characters.
slotCapacity

string (int64 format)

Baseline slots available to this reservation. A slot is a unit of computational power in BigQuery, and serves as the unit of parallelism.

Queries using this reservation might use more slots during runtime if ignoreIdleSlots is set to false, or autoscaling is enabled.

The total slotCapacity of the reservation and its siblings may exceed the total slotCount of capacity commitments. In that case, the exceeding slots will be charged with the autoscale SKU. You can increase the number of baseline slots in a reservation every few minutes. If you want to decrease your baseline slots, you are limited to once an hour if you have recently changed your baseline slot capacity and your baseline slots exceed your committed slots. Otherwise, you can decrease your baseline slots every few minutes.
ignoreIdleSlots

boolean

If false, any query or pipeline job using this reservation will use idle slots from other reservations within the same admin project. If true, a query or pipeline job using this reservation will execute with the slot capacity specified in the slotCapacity field at most.
autoscale

object (Autoscale)

The configuration parameters for the auto scaling feature.
concurrency

string (int64 format)

Job concurrency target which sets a soft upper bound on the number of jobs that can run concurrently in this reservation. This is a soft target due to asynchronous nature of the system and various optimizations for small queries. Default value is 0 which means that concurrency target will be automatically computed by the system. NOTE: this field is exposed as target job concurrency in the Information Schema, DDL and BigQuery CLI.
creationTime

string (Timestamp format)

Output only. Creation time of the reservation.

Uses RFC 3339, where generated output will always be Z-normalized and uses 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
updateTime

string (Timestamp format)

Output only. Last update time of the reservation.

Uses RFC 3339, where generated output will always be Z-normalized and uses 0, 3, 6 or 9 fractional digits. Offsets other than "Z" are also accepted. Examples: "2014-10-02T15:01:23Z", "2014-10-02T15:01:23.045123456Z" or "2014-10-02T15:01:23+05:30".
edition

enum (Edition)

Edition of the reservation.
primaryLocation

string

Output only. The current location of the reservation's primary replica. This field is only set for reservations using the managed disaster recovery feature.
secondaryLocation

string

Optional. The current location of the reservation's secondary replica. This field is only set for reservations using the managed disaster recovery feature. Users can set this in create reservation calls to create a failover reservation or in update reservation calls to convert a non-failover reservation to a failover reservation(or vice versa).
originalPrimaryLocation

string

Output only. The location where the reservation was originally created. This is set only during the failover reservation's creation. All billing charges for the failover reservation will be applied to this location.
scalingMode

enum (ScalingMode)

The scaling mode for the reservation. If the field is present but maxSlots is not present, requests will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.
maxSlots

string (int64 format)

Optional. The overall max slots for the reservation, covering slotCapacity (baseline), idle slots (if ignoreIdleSlots is false) and scaled slots. If present, the reservation won't use more than the specified number of slots, even if there is demand and supply (from idle slots). NOTE: capping a reservation's idle slot usage is best effort and its usage may exceed the maxSlots value. However, in terms of autoscale.current_slots (which accounts for the additional added slots), it will never exceed the maxSlots - baseline.

This field must be set together with the scalingMode enum value, otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.

If the maxSlots and scalingMode are set, the autoscale or autoscale.max_slots field must be unset. Otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT. However, the autoscale field may still be in the output. The autopscale.max_slots will always show as 0 and the autoscaler.current_slots will represent the current slots from autoscaler excluding idle slots. For example, if the maxSlots is 1000 and scalingMode is AUTOSCALE_ONLY, then in the output, the autoscaler.max_slots will be 0 and the autoscaler.current_slots may be any value between 0 and 1000.

If the maxSlots is 1000, scalingMode is ALL_SLOTS, the baseline is 100 and idle slots usage is 200, then in the output, the autoscaler.max_slots will be 0 and the autoscaler.current_slots will not be higher than 700.

If the maxSlots is 1000, scalingMode is IDLE_SLOTS_ONLY, then in the output, the autoscaler field will be null.

If the maxSlots and scalingMode are set, then the ignoreIdleSlots field must be aligned with the scalingMode enum value.(See details in ScalingMode comments). Otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.

Please note, the maxSlots is for user to manage the part of slots greater than the baseline. Therefore, we don't allow users to set maxSlots smaller or equal to the baseline as it will not be meaningful. If the field is present and slotCapacity>=maxSlots, requests will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.

Please note that if maxSlots is set to 0, we will treat it as unset. Customers can set maxSlots to 0 and set scalingMode to SCALING_MODE_UNSPECIFIED to disable the maxSlots feature.

Autoscale

Auto scaling settings.

JSON representation 
{
  "currentSlots": string,
  "maxSlots": string
}
Fields
currentSlots

string (int64 format)

Output only. The slot capacity added to this reservation when autoscale happens. Will be between [0, maxSlots]. Note: after users reduce maxSlots, it may take a while before it can be propagated, so currentSlots may stay in the original value and could be larger than maxSlots for that brief period (less than one minute)
maxSlots

string (int64 format)

Number of slots to be scaled when needed.

ScalingMode

The scaling mode for the reservation. This enum determines how the reservation scales up and down.

Enums
SCALING_MODE_UNSPECIFIED Default value of ScalingMode.
AUTOSCALE_ONLY

The reservation will scale up only using slots from autoscaling. It will not use any idle slots even if there may be some available. The upper limit that autoscaling can scale up to will be maxSlots - baseline. For example, if maxSlots is 1000, baseline is 200 and customer sets ScalingMode to AUTOSCALE_ONLY, then autoscalerg will scale up to 800 slots and no idle slots will be used.

Please note, in this mode, the ignoreIdleSlots field must be set to true. Otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.
IDLE_SLOTS_ONLY

The reservation will scale up using only idle slots contributed by other reservations or from unassigned commitments. If no idle slots are available it will not scale up further. If the idle slots which it is using are reclaimed by the contributing reservation(s) it may be forced to scale down. The max idle slots the reservation can be maxSlots - baseline capacity. For example, if maxSlots is 1000, baseline is 200 and customer sets ScalingMode to IDLE_SLOTS_ONLY, 1. if there are 1000 idle slots available in other reservations, the reservation will scale up to 1000 slots with 200 baseline and 800 idle slots. 2. if there are 500 idle slots available in other reservations, the reservation will scale up to 700 slots with 200 baseline and 300 idle slots. Please note, in this mode, the reservation might not be able to scale up to maxSlots.

Please note, in this mode, the ignoreIdleSlots field must be set to false. Otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.
ALL_SLOTS

The reservation will scale up using all slots available to it. It will use idle slots contributed by other reservations or from unassigned commitments first. If no idle slots are available it will scale up using autoscaling. For example, if maxSlots is 1000, baseline is 200 and customer sets ScalingMode to ALL_SLOTS, 1. if there are 800 idle slots available in other reservations, the reservation will scale up to 1000 slots with 200 baseline and 800 idle slots. 2. if there are 500 idle slots available in other reservations, the reservation will scale up to 1000 slots with 200 baseline, 500 idle slots and 300 autoscaling slots. 3. if there are no idle slots available in other reservations, it will scale up to 1000 slots with 200 baseline and 800 autoscaling slots.

Please note, in this mode, the ignoreIdleSlots field must be set to false. Otherwise the request will be rejected with error code google.rpc.Code.INVALID_ARGUMENT.

Methods

create

 Creates a new reservation resource.

delete

 Deletes a reservation.

failoverReservation

 Fail over a reservation to the secondary location.

get

 Returns information about the reservation.

list

 Lists all the reservations for the project in the specified location.

patch

 Updates an existing reservation resource.