2019/07/08 - AWS Cost Explorer Service - 1 new api methods
Changes This release introduces a new operation called GetUsageForecast, which allows you to programmatically access AWS Cost Explorer's forecasting engine on usage data (running hours, data transfer, etc).
Retrieves a forecast for how much Amazon Web Services predicts that you will use over the forecast time period that you select, based on your past usage.
See also: AWS API Documentation
Request Syntax
client.get_usage_forecast(
TimePeriod={
'Start': 'string',
'End': 'string'
},
Metric='BLENDED_COST'|'UNBLENDED_COST'|'AMORTIZED_COST'|'NET_UNBLENDED_COST'|'NET_AMORTIZED_COST'|'USAGE_QUANTITY'|'NORMALIZED_USAGE_AMOUNT',
Granularity='DAILY'|'MONTHLY'|'HOURLY',
Filter={
'Or': [
{'... recursive ...'},
],
'And': [
{'... recursive ...'},
],
'Not': {'... recursive ...'},
'Dimensions': {
'Key': 'AZ'|'INSTANCE_TYPE'|'LINKED_ACCOUNT'|'OPERATION'|'PURCHASE_TYPE'|'REGION'|'SERVICE'|'USAGE_TYPE'|'USAGE_TYPE_GROUP'|'RECORD_TYPE'|'OPERATING_SYSTEM'|'TENANCY'|'SCOPE'|'PLATFORM'|'SUBSCRIPTION_ID'|'LEGAL_ENTITY_NAME'|'DEPLOYMENT_OPTION'|'DATABASE_ENGINE'|'CACHE_ENGINE'|'INSTANCE_TYPE_FAMILY'|'BILLING_ENTITY'|'RESERVATION_ID',
'Values': [
'string',
]
},
'Tags': {
'Key': 'string',
'Values': [
'string',
]
}
},
PredictionIntervalLevel=123
)
dict
[REQUIRED]
The start and end dates of the period that you want to retrieve usage forecast for. The start date is inclusive, but the end date is exclusive. For example, if start is 2017-01-01 and end is 2017-05-01 , then the cost and usage data is retrieved from 2017-01-01 up to and including 2017-04-30 but not including 2017-05-01 .
Start (string) -- [REQUIRED]
The beginning of the time period that you want the usage and costs for. The start date is inclusive. For example, if start is 2017-01-01 , AWS retrieves cost and usage data starting at 2017-01-01 up to the end date.
End (string) -- [REQUIRED]
The end of the time period that you want the usage and costs for. The end date is exclusive. For example, if end is 2017-05-01 , AWS retrieves cost and usage data from the start date up to, but not including, 2017-05-01 .
string
[REQUIRED]
Which metric Cost Explorer uses to create your forecast.
Valid values for a GetUsageForecast call are the following:
USAGE_QUANTITY
NORMALIZED_USAGE_AMOUNT
string
[REQUIRED]
How granular you want the forecast to be. You can get 3 months of DAILY forecasts or 12 months of MONTHLY forecasts.
The GetUsageForecast operation supports only DAILY and MONTHLY granularities.
dict
The filters that you want to use to filter your forecast. Cost Explorer API supports all of the Cost Explorer filters.
Or (list) --
Return results that match either Dimension object.
(dict) --
Use Expression to filter by cost or by usage. There are two patterns:
Simple dimension values - You can set the dimension name and values for the filters that you plan to use. For example, you can filter for INSTANCE_TYPE==m4.xlarge OR INSTANCE_TYPE==c4.large . The Expression for that looks like this: { "Dimensions": { "Key": "INSTANCE_TYPE", "Values": [ "m4.xlarge", “c4.large” ] } } The list of dimension values are OR'd together to retrieve cost or usage data. You can create Expression and DimensionValues objects using either with* methods or set* methods in multiple lines.
Compound dimension values with logical operations - You can use multiple Expression types and the logical operators AND/OR/NOT to create a list of one or more Expression objects. This allows you to filter on more advanced options. For example, you can filter on ((INSTANCE_TYPE == m4.large OR INSTANCE_TYPE == m3.large) OR (TAG.Type == Type1)) AND (USAGE_TYPE != DataTransfer) . The Expression for that looks like this: { "And": [ {"Or": [ {"Dimensions": { "Key": "INSTANCE_TYPE", "Values": [ "m4.x.large", "c4.large" ] }}, {"Tags": { "Key": "TagName", "Values": ["Value1"] } } ]}, {"Not": {"Dimensions": { "Key": "USAGE_TYPE", "Values": ["DataTransfer"] }}} ] }
Note
Because each Expression can have only one operator, the service returns an error if more than one is specified. The following example shows an Expression object that creates an error.
{ "And": [ ... ], "DimensionValues": { "Dimension": "USAGE_TYPE", "Values": [ "DataTransfer" ] } }
And (list) --
Return results that match both Dimension objects.
(dict) --
Use Expression to filter by cost or by usage. There are two patterns:
Simple dimension values - You can set the dimension name and values for the filters that you plan to use. For example, you can filter for INSTANCE_TYPE==m4.xlarge OR INSTANCE_TYPE==c4.large . The Expression for that looks like this: { "Dimensions": { "Key": "INSTANCE_TYPE", "Values": [ "m4.xlarge", “c4.large” ] } } The list of dimension values are OR'd together to retrieve cost or usage data. You can create Expression and DimensionValues objects using either with* methods or set* methods in multiple lines.
Compound dimension values with logical operations - You can use multiple Expression types and the logical operators AND/OR/NOT to create a list of one or more Expression objects. This allows you to filter on more advanced options. For example, you can filter on ((INSTANCE_TYPE == m4.large OR INSTANCE_TYPE == m3.large) OR (TAG.Type == Type1)) AND (USAGE_TYPE != DataTransfer) . The Expression for that looks like this: { "And": [ {"Or": [ {"Dimensions": { "Key": "INSTANCE_TYPE", "Values": [ "m4.x.large", "c4.large" ] }}, {"Tags": { "Key": "TagName", "Values": ["Value1"] } } ]}, {"Not": {"Dimensions": { "Key": "USAGE_TYPE", "Values": ["DataTransfer"] }}} ] }
Note
Because each Expression can have only one operator, the service returns an error if more than one is specified. The following example shows an Expression object that creates an error.
{ "And": [ ... ], "DimensionValues": { "Dimension": "USAGE_TYPE", "Values": [ "DataTransfer" ] } }
Not (dict) --
Return results that don't match a Dimension object.
Dimensions (dict) --
The specific Dimension to use for Expression .
Key (string) --
The names of the metadata types that you can use to filter and group your results. For example, AZ returns a list of Availability Zones.
Values (list) --
The metadata values that you can use to filter and group your results. You can use GetDimensionValues to find specific values.
Valid values for the SERVICE dimension are Amazon Elastic Compute Cloud - Compute , Amazon Elasticsearch Service , Amazon ElastiCache , Amazon Redshift , and Amazon Relational Database Service .
(string) --
Tags (dict) --
The specific Tag to use for Expression .
Key (string) --
The key for the tag.
Values (list) --
The specific value of the tag.
(string) --
integer
Cost Explorer always returns the mean forecast as a single point. You can request a prediction interval around the mean by specifying a confidence level. The higher the confidence level, the more confident Cost Explorer is about the actual value falling in the prediction interval. Higher confidence levels result in wider prediction intervals.
dict
Response Syntax
{
'Total': {
'Amount': 'string',
'Unit': 'string'
},
'ForecastResultsByTime': [
{
'TimePeriod': {
'Start': 'string',
'End': 'string'
},
'MeanValue': 'string',
'PredictionIntervalLowerBound': 'string',
'PredictionIntervalUpperBound': 'string'
},
]
}
Response Structure
(dict) --
Total (dict) --
How much you're forecasted to use over the forecast period.
Amount (string) --
The actual number that represents the metric.
Unit (string) --
The unit that the metric is given in.
ForecastResultsByTime (list) --
The forecasts for your query, in order. For DAILY forecasts, this is a list of days. For MONTHLY forecasts, this is a list of months.
(dict) --
The forecast created for your query.
TimePeriod (dict) --
The period of time that the forecast covers.
Start (string) --
The beginning of the time period that you want the usage and costs for. The start date is inclusive. For example, if start is 2017-01-01 , AWS retrieves cost and usage data starting at 2017-01-01 up to the end date.
End (string) --
The end of the time period that you want the usage and costs for. The end date is exclusive. For example, if end is 2017-05-01 , AWS retrieves cost and usage data from the start date up to, but not including, 2017-05-01 .
MeanValue (string) --
The mean value of the forecast.
PredictionIntervalLowerBound (string) --
The lower limit for the prediction interval.
PredictionIntervalUpperBound (string) --
The upper limit for the prediction interval.