Skip to content

Latest commit

 

History

History
357 lines (301 loc) · 18.4 KB

README.md

File metadata and controls

357 lines (301 loc) · 18.4 KB
Raw

Autoscaler tool for Cloud Spanner

Autoscaler

Retrieve metrics for one or more Cloud Spanner Instances
Home · Poller component · Scaler component · Forwarder component · Terraform configuration · Monitoring

Table of Contents

Overview

The Poller component takes an array of Cloud Spanner instances and obtains load metrics for each of them from Cloud Monitoring. This array may come from the payload of a Cloud PubSub message or from configuration held in a Kubernetes ConfigMap, depending on configuration.

Then for each Spanner instance it publishes a message via the specified Cloud PubSub topic or via HTTP, which includes the metrics and part of the configuration for the Spanner instance.

The Scaler component will receive the message, compare the metric values with the recommended thresholds, plus or minus an allowed margin, and if any of the values fall outside of this range, the Scaler component will adjust the number of nodes in the Spanner instance accordingly. Note that the thresholds are different depending if a Spanner instance is regional or multi-region.

Configuration parameters

The following are the configuration parameters consumed by the Poller component. Some of these parameters are forwarded to the Scaler component as well.

In the case of the Poller and Scaler components deployed to Cloud Functions, the parameters are defined using JSON in the payload of the PubSub message that is published by the Cloud Scheduler job. When deployed to Kubernetes, the configuration parameters are defined in YAML in a Kubernetes ConfigMap.

See the configuration section in the home page for instructions on how to change the payload.

Required

Key Description
projectId Project ID of the Cloud Spanner to be monitored by the Autoscaler
instanceId Instance ID of the Cloud Spanner to be monitored by the Autoscaler

Required for a Cloud Functions deployment

Key Description
scalerPubSubTopic PubSub topic for the Poller function to publish messages for the Scaler function. The topic must be in the format projects/{projects}/topics/{topicId}.

Optional

Key Default Value Description
units NODES Specifies the units that capacity will be measured in NODES or PROCESSING_UNITS.
minSize 1 N or 100 PU Minimum number of Cloud Spanner nodes or processing units that the instance can be scaled IN to.　Do not include the unit (N or PU) in the value.
maxSize 3 N or 2000 PU Maximum number of Cloud Spanner nodes or processing units that the instance can be scaled OUT to.　Do not include the unit (N or PU) in the value.
scalingMethod STEPWISE Scaling method that should be used. Options are: STEPWISE, LINEAR, DIRECT. See the scaling methods section in the Scaler component page for more information.
stepSize 2 N or 200 PU Number of nodes that should be added or removed when scaling with the STEPWISE method. When the Spanner instance size is over 1000 PUs, scaling will be done in steps of 1000 PUs. For more information see the Spanner compute capacity documentation.　Do not include the unit (N or PU) in the value.
overloadStepSize 5 N or 500 PU Number of nodes that should be added when the Cloud Spanner instance is overloaded, and the STEPWISE method is used.　Do not include the unit (N or PU) in the value.
scaleOutCoolingMinutes 5 Minutes to wait after scaling IN or OUT before a scale OUT event can be processed.
scaleInCoolingMinutes 30 Minutes to wait after scaling IN or OUT before a scale IN event can be processed.
overloadCoolingMinutes 5 Minutes to wait after scaling IN or OUT before a scale OUT event can be processed, when the Spanner instance is overloaded. An instance is overloaded if its High Priority CPU utilization is over 90%.
stateProjectId ${projectId} The project ID where the Autoscaler state will be persisted. By default it is persisted using Cloud Firestore in the same project as the Spanner instance.
stateDatabase Object An Object that can override the database for managing the state of the Autoscaler. The default database is Firestore. Refer to the state database for details.
metrics Array Array of objects that can override the values in the metrics used to decide when the Cloud Spanner instance should be scaled IN or OUT. Refer to the metrics definition table to see the fields used for defining metrics.
scaleInLimit undefined Percentage (integer) of the total instance size that can be removed in a scale in event when using the linear algorithm. For example if set to 20, only 20% of the instance size can be removed in a single scaling event, when scaleInLimit is undefined a limit is not enforced.
minNodes (DEPRECATED) 1 DEPRECATED: Minimum number of Cloud Spanner nodes that the instance can be scaled IN to.
maxNodes (DEPRECATED) 3 DEPRECATED: Maximum number of Cloud Spanner nodes that the instance can be scaled OUT to.
downstreamPubSubTopic undefined Set this parameter to projects/${projectId}/topics/downstream-topic if you want the the Autoscaler to publish events that can be consumed by downstream applications. See Downstream messaging for more information.
scalerURL http://scaler URL where the scaler service receives HTTP requests.

Metrics parameters

The table describes the objects used to define metrics. These can be provided in the configuration objects to customize the metrics used to autoscale your Cloud Spanner instances.

To specify a custom threshold specify the name of the metrics to customize followed by the parameter values you wish to change. The updated parameters will be merged with the default metric parameters.

Selectors

Key Description
name A unique name of the for the metric to be evaulated. If you want to override the default metrics, their names are: high_priority_cpu, rolling_24_hr and storage.

Parameters

When defining a metric for the Autoscaler there are two key components: thresholds and a Cloud Monitoring time series metric comprised of a filter, reducer, aligner and period. Having a properly defined metric is critical to the opertional of the Autoscaler, please refer to Filtering and aggregation: manipulating time series for a complete discussion on building metric filters and aggregating data points.

Key Default Description
filter The Cloud Spanner metric and filter that should be used when querying for data. The Autoscaler will automatically add the filter expressions for Spanner instance resources, instance id and project id.
reducer REDUCE_SUM The reducer specifies how the data points should be aggregated when querying for metrics, typically REDUCE_SUM. For more details please refer to Alert Policies - Reducer documentation.
aligner ALIGN_MAX The aligner specifies how the data points should be aligned in the time series, typically ALIGN_MAX. For more details please refer to Alert Policies - Aligner documentation.
period 60 Defines the period of time in units of seconds at which aggregation takes place. Typically the period should be 60.
regional_threshold Threshold used to evaluate if a regional instance needs to be scaled in or out.
multi_regional_threshold Threshold used to evaluate if a multi-regional instance needs to be scaled in or out.
regional_margin 5 Margin above and below the threshold where the metric value is allowed. If the metric falls outside of the range [threshold - margin, threshold + margin], then the regional instance needs to be scaled in or out.
multi_regional_margin 5 Margin above and below the threshold where the metric value is allowed. If the metric falls outside of the range [threshold - margin, threshold + margin], then the multi regional instance needs to be scaled in or out.

Custom metrics, thresholds and margins

The Autoscaler determines the number of nodes or processing units to be added or substracted to an instance based on the Spanner recommended thresholds for High Priority CPU, 24 hour rolling average CPU and Storage utilization metrics.

Google recommends using the provided metrics, thresholds and margins unchanged. However, in some cases you may want to modify these or use a custom metric, for example: if reaching the default upper limit triggers an alert to your operations team, you could make the Autoscaler react to a more conservative threshold to avoid alerts being triggered.

Thresholds

To modify the recommended thresholds, add the metrics parameter to your configuration and specify name (high_priority_cpu, rolling_24_hr and storage) of the metric to be changed and desired regional_threshold or multi_regional_threshold for your Cloud Spanner instance.

Margins

A margin defines an upper and a lower limit around the threshold. An autoscaling event will be triggered only if the metric value falls above the upper limit, or below the lower limit.

The objective of this parameter is to avoid autoscaling events being triggered for small workload fluctuations around the threshold, thus creating a smoothing effect in autoscaler actions. The threshold and metric together define a range [threshold - margin, threshold + margin], where the metric value is allowed. The smaller the margin, the narrower the range, resulting in higher probability that an autoscaling event is triggered.

By default, the margin value is 5 for both regional and multi-regional instances. You can change the default value by specifying regional_margin or multi_regional_margin in the metric parameters. Specifying a margin parameter for a metric is optional.

Metrics

To create a custom metric, add the metrics parameter to your configuration specifying the required fields (name, filter, regional_threshold, multi_regional_threshold). The period, reducer and aligner are defaulted but can also be specified in the metric definition.

The Cloud Spanner documentation contains details for the Cloud Spanner metric and filter that should be used when querying for data. The Autoscaler will automatically add the filter expressions for Spanner instance resources, instance id and project id, unless you have chosen a name for your custom metric that matches one of the default metrics, in which case you may either:

  1. Choose a different name for your custom metric (recommended), or
  2. Construct the full filter expression manually to include the Spanner details and project id.

State Database

The table describes the objects used to specify the database for managing the state of the Autoscaler.

Key Default Description
name firestore Name of the database for managing the state of the Autoscaler. By default, Firestore is used. The currently supported values are firestore and spanner.

State Managing in Cloud Spanner

If the value of name is spanner, the following values are required.

Key Description
instanceId The instance id of Cloud Spanner which you want to manage the state.
databaseId The database id of Cloud Spanner instance which you want to manage the state.

When using Cloud Spanner to manage the state, a table with the following DDL is created at runtime.

CREATE TABLE spannerAutoscaler (
  id STRING(MAX),
  lastScalingTimestamp TIMESTAMP,
  createdOn TIMESTAMP,
  updatedOn TIMESTAMP,
  lastScalingCompleteTimestamp TIMESTAMP,
  scalingOperationId STRING(MAX),
  scalingRequestedSize INT64,
  scalingMethod STRING(MAX),
  scalingPreviousSize INT64,
) PRIMARY KEY (id)

Note: If you are upgrading from v1.x, then you need to add the 5 new columns to the spanner schema using the following DDL statements

ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS lastScalingCompleteTimestamp TIMESTAMP;
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingOperationId STRING(MAX);
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingRequestedSize INT64;
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingMethod STRING(MAX);
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingPreviousSize INT64;

Note: If you are upgrading from V2.0.x, then you need to add the 3 new columns to the spanner schema using the following DDL statements

ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingRequestedSize INT64;
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingMethod STRING(MAX);
ALTER TABLE spannerAutoscaler ADD COLUMN IF NOT EXISTS scalingPreviousSize INT64;

Example configuration for Cloud Functions

[
    {
        "projectId": "basic-configuration",
        "instanceId": "another-spanner1",
        "scalerPubSubTopic": "projects/my-spanner-project/topics/spanner-scaling",
        "units": "NODES",
        "minSize": 5,
        "maxSize": 30,
        "scalingMethod": "DIRECT"
    },{
        "projectId": "custom-threshold",
        "instanceId": "spanner1",
        "scalerPubSubTopic": "projects/my-spanner-project/topics/spanner-scaling",
        "units": "PROCESSING_UNITS",
        "minSize": 100,
        "maxSize": 3000,
        "metrics": [
          {
            "name": "high_priority_cpu",
            "regional_threshold": 40,
            "regional_margin": 3
          }
        ]
    },{
        "projectId": "custom-metric",
        "instanceId": "another-spanner1",
        "scalerPubSubTopic": "projects/my-spanner-project/topics/spanner-scaling",
        "units": "NODES",
        "minSize": 5,
        "maxSize": 30,
        "scalingMethod": "LINEAR",
        "scaleInLimit": 25,
        "metrics": [
          {
            "name": "my_custom_metric",
            "filter": "metric.type=\"spanner.googleapis.com/instance/resource/metric\"",
            "regional_threshold": 40,
            "multi_regional_threshold": 30
          }
        ]
    }
]

Example configuration for Google Kubernetes Engine

apiVersion: v1
kind: ConfigMap
metadata:
  name: autoscaler-config
  namespace: spanner-autoscaler
data:
  autoscaler-config.yaml: |
    ---
    - projectId: spanner-autoscaler-test
      instanceId: spanner-scaling-direct
      units: NODES
      minSize: 5
      maxSize: 30
      scalingMethod: DIRECT
    - projectId: spanner-autoscaler-test
      instanceId: spanner-scaling-threshold
      units: PROCESSING_UNITS
      minSize: 100
      maxSize: 3000
      metrics:
      - name: high_priority_cpu
        regional_threshold: 40
        regional_margin: 3
    - projectId: spanner-autoscaler-test
      instanceId: spanner-scaling-custom
      units: NODES
      minSize: 5
      maxSize: 30
      scalingMethod: LINEAR
      scaleInLimit: 25
      metrics:
      - name: my_custom_metric
        filter: metric.type="spanner.googleapis.com/instance/resource/metric"
        regional_threshold: 40
        multi_regional_threshold: 30