HashiConf 2024 Now streaming live from Boston! Attend for free
Terraform
AWS Services

Manage AWS DynamoDB scale

  • 35min
  • |
  • Terraform

AWS DynamoDB is a fully managed, serverless, key-value NoSQL database that supports a wide variety of use cases. AWS provides many configuration options to manage your DynamoDB tables' capacity and performance.

In this tutorial, you will learn how to configure these options with Terraform. First you will provision a DynamoDB table. Next you will use Terraform to configure features that help you manage DynamoDB scale and capacity. Then you will use Terraform to load data into your table, and query the data with the AWS CLI.

Prerequisites

You can complete this tutorial using the same workflow with either Terraform Community Edition or HCP Terraform. HCP Terraform is a platform that you can use to manage and execute your Terraform projects. It includes features like remote state and execution, structured plan output, workspace resource summaries, and more.

Select the Terraform Community Edition tab to complete this tutorial using Terraform Community Edition.

This tutorial assumes that you are familiar with the Terraform and HCP Terraform workflows. If you are new to Terraform, complete the [Get Started tutorials(/collections/terraform/aws-get-started) first. If you are new to HCP Terraform, complete the HCP Terraform Get Started tutorials first.

In order to complete this tutorial, you will need the following:

Note

Some of the infrastructure in this tutorial does not qualify for the AWS free tier. Destroy the infrastructure at the end of the tutorial to avoid unnecessary charges. We are not responsible for any charges that you incur.

Clone the example repository

Clone the example repository for this tutorial, which contains Terraform configuration for a DynamoDB table.

$ git clone https://github.com/hashicorp-education/learn-terraform-aws-dynamodb-scale

Change into the repository directory.

$ cd learn-terraform-aws-dynamodb-scale

Review the configuration

The example configuration defines a DynamoDB table configured to store environmental data such as temperature, pressure, and humidity. Devices such as weather stations or environmental sensors often report this kind of data. The example data, found in data/example_environments.csv, consists of data for the following table attributes:

  • userId: A unique identifier for the owner of the device.
  • deviceId: A unique identifier for the device.
  • eventId: A unique identifier for a given event.
  • geoLocation: The geographical location of the device.
  • epochS: The time, in epoch seconds, when the device recorded the event.
  • expiry: The time, in epoch seconds, past which an event no longer needs to be stored in your DynamoDB table.
  • tempC: The temperature, in celsius.
  • humidityPct: The humidity, represented as a percentage.
  • pressurePa: The atmospheric pressure, in Pascals.

Open main.tf to review the initial configuration, which defines a DynamoDB table with a randomly generated name.

main.tf

resource "random_pet" "table_name" {
  prefix    = "environment"
  separator = "_"
  length    = 4
}

resource "aws_dynamodb_table" "environment" {
  name         = random_pet.table_name.id
  billing_mode = "PAY_PER_REQUEST"

  hash_key  = "deviceId"
  range_key = "epochS"

Note

AWS uses different terminology for some DynamoDB concepts in their documentation and the API. The Terraform AWS provider uses the API term and so does this tutorial. We include the documentation name in parentheses. For example: "hash (partition) key".

All DynamoDB tables require a name and a primary key. The random_pet.table_name resource provides a unique name for the table. This configuration uses a composite primary key consisting of a hash (partition) key, deviceId, and a range (sort) key, epochS.

The billing_mode argument configures the table to use the pay-per-request billing (capacity) mode, which allows AWS to set read and write capacity for you. The pay-per-request mode can be more cost-effective than provisioned capacity for some unpredictable workloads.

main.tf

  attribute {
    name = "deviceId"
    type = "S"
  }

  attribute {
    name = "epochS"
    type = "N"
  }
}

The example configuration sets attributes for the hash key, deviceId, and range key, epochS. Only configure attribute blocks for table attributes that you use in the keys or other indexes. You can store other attributes in your DynamoDB table, but the AWS API will error if you define them in your Terraform configuration. For example, the table will store values for tempC, but there is no attribute block for it in your configuration because your table does not use that attribute in an index. If you were to include an attribute block for tempC, AWS would return an error:

│ Error: 1 error occurred:
│   * all attributes must be indexed. Unused attributes: ["tempC"]

The configuration in outputs.tf defines two outputs for the table's name and Amazon Resource Name (ARN).

outputs.tf

output "environment_table_name" {
  description = "Name of the environment DynamoDB table"
  value       = aws_dynamodb_table.environment.name
}

output "environment_table_arn" {
  description = "ARN of the environment DynamoDB table"
  value       = aws_dynamodb_table.environment.arn
}

Provision a minimal DynamoDB table

Set the TF_CLOUD_ORGANIZATION environment variable to your HCP Terraform organization name. This will configure your HCP Terraform integration.

$ export TF_CLOUD_ORGANIZATION=

Initialize your configuration. Terraform will automatically create the learn-terraform-aws-dynamodb-scale workspace in your HCP Terraform organization.

$ terraform init

Initializing HCP Terraform...

Initializing provider plugins...
- Reusing previous version of hashicorp/aws from the dependency lock file
- Reusing previous version of hashicorp/random from the dependency lock file
- Installing hashicorp/aws v4.21.0...
- Installed hashicorp/aws v4.21.0 (signed by HashiCorp)
- Installing hashicorp/random v3.2.0...
- Installed hashicorp/random v3.2.0 (signed by HashiCorp)

HCP Terraform has been successfully initialized!

You may now begin working with HCP Terraform. Try running "terraform plan" to
see any changes that are required for your infrastructure.

If you ever set or change modules or Terraform Settings, run "terraform init"
again to reinitialize your working directory.

Note

This tutorial assumes that you are using a tutorial-specific HCP Terraform organization with a global variable set of your AWS credentials. Review the Create a Credential Variable Set for detailed guidance. If you are using a scoped variable set, assign it to your new workspace now.

Apply the configuration. Respond to the confirmation prompt with a yes to create your DynamoDB table.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-sjJiEKFWX9HNRaNs

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  + create

Terraform will perform the following actions:

  # aws_dynamodb_table.environment will be created
  + resource "aws_dynamodb_table" "environment" {

##...

Plan: 2 to add, 0 to change, 0 to destroy.

Changes to Outputs:
  + environment_table_arn  = (known after apply)
  + environment_table_name = (known after apply)

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

random_pet.table_name: Creating...
random_pet.table_name: Creation complete after 0s [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Creating...
aws_dynamodb_table.environment: Creation complete after 6s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 2 added, 0 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Tip

This tutorial shows the output for Terraform commands run in HCP Terraform. If you are following the Terraform Community Edition workflow, the output will differ slightly, but the results will be the same.

Scale your DynamoDB table

In this tutorial, you will configure several options to manage your table's capacity and performance.

  1. Configure provisioned capacity
  2. Configure Autoscaling
  3. Add secondary indexes
  4. Configure global tables
  5. Manage TTL
  6. Change the table class

You can configure most of these options independently from the others, except as noted below.

Configure provisioned capacity

The example configuration uses the pay-per-request billing (capacity) mode. Provisioned capacity can be more cost effective, especially for predictable workloads. When you configure provisioned capacity, you must specify the read and write capacity for your table. For example, you may determine that your application requires a read capacity of 5, and a write capacity of 2.

Add the following variable definitions to variables.tf to configure your table's read and write capacity.

variables.tf

variable "env_table_read_capacity" {
  description = "Read capacity for environment table."
  type        = number
  default     = 5
}

variable "env_table_write_capacity" {
  description = "Write capacity for environment table."
  type        = number
  default     = 2
}

Next, update your table to use provisioned billing (capacity) mode, and configure the read and write capacity with those variables.

main.tf

resource "aws_dynamodb_table" "environment" {
  name         = "environment"
  billing_mode = "PROVISIONED"

  read_capacity  = var.env_table_read_capacity
  write_capacity = var.env_table_write_capacity

  hash_key  = "deviceId"
  range_key = "epochS"
##...

Apply this change to update your table's billing mode. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-oKS4np8PyAo5gj5d

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Refreshing state... [id=environment_partially_clearly_polished_moth]

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_dynamodb_table.environment will be updated in-place
  ~ resource "aws_dynamodb_table" "environment" {
      ~ billing_mode   = "PAY_PER_REQUEST" -> "PROVISIONED"
        id             = "environment_partially_clearly_polished_moth"
        name           = "environment_partially_clearly_polished_moth"
      ~ read_capacity  = 0 -> 5
        tags           = {}
      ~ write_capacity = 0 -> 2
        # (5 unchanged attributes hidden)

        # (4 unchanged blocks hidden)
    }

Plan: 0 to add, 1 to change, 0 to destroy.

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table.environment: Modifying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Modifications complete after 6s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 0 added, 1 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Changing the billing (capacity) mode or updating provisioned read and write capacity does not require recreating the table, so Terraform updated it in place.

AWS lets you change a table's billing mode once every 24 hours. If you tried to change it again within this time window, AWS would report an error:

│ Error: error updating DynamoDB Table (environment_partially_clearly_polished_moth):
LimitExceededException: Subscriber limit exceeded: Update to PayPerRequest mode
are limited to once in 1 day(s). Last update at Fri Jul 08 17:42:44 UTC 2022.
Next update can be made at Sat Jul 09 17:42:44 UTC 2022

Configure autoscaling

When you configure provisioned capacity for your table, you can also configure Application Autoscaling to have AWS manage the table's read and write capacity. When you do so, AWS will update your table's read and write capacity according to the autoscaling rules you define. Autoscaling will increase your table's capacity when more traffic is incoming, and reduce it after traffic subsides.

Before you enable autoscaling, instruct Terraform to ignore changes to your table's read and write capacity attributes. Otherwise, whenever you run terraform apply in the future, Terraform will report that these values have changed outside of Terraform's control, and revert them to the original values.

In main.tf, add a lifecycle block to your your table's configuration so Terraform will ignore changes to read_capacity and write_capacity.

main.tf

    attribute {
    name = "epochS"
    type = "N"
  }

  lifecycle {
    ignore_changes = [read_capacity, write_capacity]
  }
}

Also in main.tf, add read and write policies targeting your table.

main.tf

resource "aws_appautoscaling_target" "environment_table_read_target" {
  max_capacity       = 20
  min_capacity       = 5
  resource_id        = "table/${aws_dynamodb_table.environment.name}"
  scalable_dimension = "dynamodb:table:ReadCapacityUnits"
  service_namespace  = "dynamodb"
}

resource "aws_appautoscaling_policy" "environment_table_read_policy" {
  name               = "DynamoDBReadCapacityUtilization:${aws_appautoscaling_target.environment_table_read_target.resource_id}"
  policy_type        = "TargetTrackingScaling"
  resource_id        = aws_appautoscaling_target.environment_table_read_target.resource_id
  scalable_dimension = aws_appautoscaling_target.environment_table_read_target.scalable_dimension
  service_namespace  = aws_appautoscaling_target.environment_table_read_target.service_namespace

  target_tracking_scaling_policy_configuration {
    predefined_metric_specification {
      predefined_metric_type = "DynamoDBReadCapacityUtilization"
    }

    target_value = 70.0
  }
}

resource "aws_appautoscaling_target" "environment_table_write_target" {
  max_capacity       = 10
  min_capacity       = 5
  resource_id        = "table/${aws_dynamodb_table.environment.name}"
  scalable_dimension = "dynamodb:table:WriteCapacityUnits"
  service_namespace  = "dynamodb"
}

resource "aws_appautoscaling_policy" "environment_table_write_policy" {
  name               = "DynamoDBWriteCapacityUtilization:${aws_appautoscaling_target.environment_table_write_target.resource_id}"
  policy_type        = "TargetTrackingScaling"
  resource_id        = aws_appautoscaling_target.environment_table_write_target.resource_id
  scalable_dimension = aws_appautoscaling_target.environment_table_write_target.scalable_dimension
  service_namespace  = aws_appautoscaling_target.environment_table_write_target.service_namespace

  target_tracking_scaling_policy_configuration {
    predefined_metric_specification {
      predefined_metric_type = "DynamoDBWriteCapacityUtilization"
    }

    target_value = 70.0
  }
}

The min_capacity and max_capacity attributes of the aws_appautoscaling_target resources configure the upper and lower bounds for your table's read and write capacities. Setting the target_value to 70.0 for each capacity type means that when your table is using less than 70 percent of its current read or write capacity, autoscaling will reduce its capacity, and when your table uses more than 70 percent of its currentt capacity, autoscaling will increase capacity.

Apply this change to configure autoscaling for your table. Terraform will ignore changes to read and write capacity parameters going forward. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-CwuBhQgdjG6BABnM

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Refreshing state... [id=environment_partially_clearly_polished_moth]

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  + create

Terraform will perform the following actions:

  # aws_appautoscaling_policy.environment_table_read_policy will be created
  + resource "aws_appautoscaling_policy" "environment_table_read_policy" {
##...

Plan: 4 to add, 0 to change, 0 to destroy.

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_appautoscaling_target.environment_table_write_target: Creating...
aws_appautoscaling_target.environment_table_read_target: Creating...
aws_appautoscaling_target.environment_table_write_target: Creation complete after 0s [id=table/environment_partially_clearly_polished_moth]
aws_appautoscaling_target.environment_table_read_target: Creation complete after 0s [id=table/environment_partially_clearly_polished_moth]
aws_appautoscaling_policy.environment_table_write_policy: Creating...
aws_appautoscaling_policy.environment_table_read_policy: Creating...
aws_appautoscaling_policy.environment_table_read_policy: Creation complete after 1s [id=DynamoDBReadCapacityUtilization:table/environment_partially_clearly_polished_moth]
aws_appautoscaling_policy.environment_table_write_policy: Creation complete after 1s [id=DynamoDBWriteCapacityUtilization:table/environment_partially_clearly_polished_moth]

Apply complete! Resources: 4 added, 0 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Adding or removing autoscaling from your DynamoDB tables does not require replacing your table, so Terraform updated it in place.

Add secondary indexes

Secondary indexes can speed up queries that do not involve the table's primary key. There are two types of secondary indexes: local, and global. Local secondary indexes use the same partition key as the table, but a different sort key, and are strongly consistent. Global secondary indexes use a different partition key, scale independently from the table, and are eventually consistent. AWS recommends using global secondary indexes unless you specifically need a Local secondary index, for example when you need strong consistency.

Add a local secondary index

Your application may often query a range of data based upon the eventId table attribute, and need strongly consistent reads for these queries. Add a local secondary index to meet this need.

In the table resource in main.tf, add the attribute that the local secondary index must query on, and the local secondary index itself.

main.tf

##...
  attribute {
    name = "epochS"
    type = "N"
  }

  attribute {
    name = "eventId"
    type = "S"
  }

  local_secondary_index {
    name            = "by_eventId"
    range_key       = "eventId"
    projection_type = "ALL"
  }

  lifecycle {
    ignore_changes = [read_capacity, write_capacity]
  }
}

The new table attribute, eventId defines the required attribute for the range key that your local secondary index uses.

Warning

Terraform destroys and recreates your table whenever you add, remove, or update a local secondary index, which deletes all the items in the table. This is because AWS requires that you define local secondary indexes when you create your table.

Apply this change to replace your table with one that includes this local secondary index. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-KLe2U9HjMijdJQfn

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_appautoscaling_target.environment_table_write_target: Refreshing state... [id=table/environment_partially_clearly_polished_moth]
aws_appautoscaling_target.environment_table_read_target: Refreshing state... [id=table/environment_partially_clearly_polished_moth]
aws_appautoscaling_policy.environment_table_read_policy: Refreshing state... [id=DynamoDBReadCapacityUtilization:table/environment_partially_clearly_polished_moth]
aws_appautoscaling_policy.environment_table_write_policy: Refreshing state... [id=DynamoDBWriteCapacityUtilization:table/environment_partially_clearly_polished_moth]

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
-/+ destroy and then create replacement

Terraform will perform the following actions:

  # aws_dynamodb_table.environment must be replaced
-/+ resource "aws_dynamodb_table" "environment" {

##...

      + local_secondary_index { # forces replacement
          + name               = "by_eventId"
          + non_key_attributes = []
          + projection_type    = "ALL"
          + range_key          = "eventId"
        }

##...

Plan: 1 to add, 0 to change, 1 to destroy.

Changes to Outputs:
  ~ environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth" -> (known after apply)

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table.environment: Destroying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Destruction complete after 6s
aws_dynamodb_table.environment: Creating...
aws_dynamodb_table.environment: Still creating... [10s elapsed]
aws_dynamodb_table.environment: Creation complete after 13s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 1 added, 0 changed, 1 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Local secondary indexes use the base table's read and write capacity, so you must account for access to the secondary index when you configure provisioned or autoscaling capacity for your table.

Add a global secondary index

For use cases that do not require strong read consistency, AWS recommends that you use global secondary indexes instead of local secondary indexes. Global secondary indexes also allow your application to query your table by a different primary key from your main table.

Add a geoLocation attribute to your table, and a global secondary index to allow users of your table to query events by geoLocation.

main.tf

##...
  attribute {
    name = "eventId"
    type = "S"
  }

  attribute {
    name = "geoLocation"
    type = "S"
  }

  local_secondary_index {
    name            = "by_eventId"
    range_key       = "eventId"
    projection_type = "ALL"
  }

  global_secondary_index {
    name               = "by_geoLocation"
    hash_key           = "geoLocation"
    range_key          = "epochS"
    write_capacity     = 5
    read_capacity      = 10
    projection_type    = "INCLUDE"
    non_key_attributes = ["userId", "location"]
  }
##...

Like keys for local secondary indexes, you must include keys for global secondary indexes as attributes in the table definition.

AWS recommends autoscaling global secondary indexes whenever you use autoscaling on the table itself in order to ensure consistent performance across your table and GSIs. Add autoscaling configuration for the global secondary index to main.tf.

main.tf

resource "aws_appautoscaling_target" "environment_table_by_geo_location_read_target" {
  max_capacity       = 20
  min_capacity       = 5
  resource_id        = "table/${aws_dynamodb_table.environment.name}/index/by_geoLocation"
  scalable_dimension = "dynamodb:index:ReadCapacityUnits"
  service_namespace  = "dynamodb"
}

resource "aws_appautoscaling_policy" "environment_table_by_geo_location_read_policy" {
  name               = "DynamoDBReadCapacityUtilization:${aws_appautoscaling_target.environment_table_by_geo_location_read_target.resource_id}"
  policy_type        = "TargetTrackingScaling"
  resource_id        = aws_appautoscaling_target.environment_table_by_geo_location_read_target.resource_id
  scalable_dimension = aws_appautoscaling_target.environment_table_by_geo_location_read_target.scalable_dimension
  service_namespace  = aws_appautoscaling_target.environment_table_by_geo_location_read_target.service_namespace

  target_tracking_scaling_policy_configuration {
    predefined_metric_specification {
      predefined_metric_type = "DynamoDBReadCapacityUtilization"
    }

    target_value = 70.0
  }
}

resource "aws_appautoscaling_target" "environment_table_by_geo_location_write_target" {
  max_capacity       = 10
  min_capacity       = 5
  resource_id        = "table/${aws_dynamodb_table.environment.name}/index/by_geoLocation"
  scalable_dimension = "dynamodb:index:WriteCapacityUnits"
  service_namespace  = "dynamodb"
}

resource "aws_appautoscaling_policy" "environment_table_by_geo_location_write_policy" {
  name               = "DynamoDBWriteCapacityUtilization:${aws_appautoscaling_target.environment_table_by_geo_location_write_target.resource_id}"
  policy_type        = "TargetTrackingScaling"
  resource_id        = aws_appautoscaling_target.environment_table_by_geo_location_write_target.resource_id
  scalable_dimension = aws_appautoscaling_target.environment_table_by_geo_location_write_target.scalable_dimension
  service_namespace  = aws_appautoscaling_target.environment_table_by_geo_location_write_target.service_namespace

  target_tracking_scaling_policy_configuration {
    predefined_metric_specification {
      predefined_metric_type = "DynamoDBWriteCapacityUtilization"
    }

    target_value = 70.0
  }
}

Apply this change to add the global secondary index and autoscaling configuration to your table. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-w6UhqQgAUsDmk5nT

Waiting for the plan to start...

##...

Terraform will perform the following actions:

  # aws_appautoscaling_policy.environment_table_by_geo_location_read_policy will be created
  + resource "aws_appautoscaling_policy" "environment_table_by_geo_location_read_policy" {

##...

Plan: 8 to add, 1 to change, 0 to destroy.

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table.environment: Modifying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 10s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 20s elapsed]

##...

aws_appautoscaling_policy.environment_table_by_geo_location_read_policy: Creation complete after 1s [id=DynamoDBReadCapacityUtilization:table/environment_partially_clearly_polished_moth/index/by_geoLocation]
aws_appautoscaling_policy.environment_table_write_policy: Creation complete after 1s [id=DynamoDBWriteCapacityUtilization:table/environment_partially_clearly_polished_moth]

Apply complete! Resources: 8 added, 1 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

It may take a few minutes for AWS to provision your global secondary index.

Note

Since the global secondary index scales automatically, whenever you apply changes to this configuration, Terraform will report that the read and write capacity of the index have changed. A limitation in the DynamoDB API prevents Terraform's AWS provider from ignoring changes to these attributes.

AWS allows you to create, update, and remove global secondary indexes independently from the main table, and the indexes can have their own primary key and read and write capacities.

Configure global tables

Global tables replicate your DynamoDB table across AWS regions. You must enable streaming on your DynamoDB table before you can use global tables, and your table must either be in the PAY_PER_REQUEST billing mode, or have autoscaling configured. The DynamoDB stream publishes updates whenever you add, delete, or modify the items in your table. Your global tables will consume updates from your stream to stay in sync with your main table.

For example, your table is provisioned in the us-east-1 region, but you may want to replicate your data to other regions to improve performance for customers in those locations. Configure global table replicas in the us-west-1 and ap-northeast-1 regions.

Add a variable to configure the regions you want AWS to replicate your DynamoDB table across.

variables.tf

variable "replica_regions" {
  description = "AWS region names (eg 'us-east-1') to create Global Table replicas in."
  type        = list(string)
  default     = []
}

Next, create a new file called terraform.tfvars and set a value for the replica_regions variable.

terraform.tfvars

replica_regions = ["us-west-1", "ap-northeast-1"]

Now configure global tables by enabling streaming and adding replica blocks to your table resource in main.tf.

main.tf

resource "aws_dynamodb_table" "environment" {
  name         = random_pet.table_name.id
  billing_mode = "PROVISIONED"

  read_capacity  = var.env_table_read_capacity
  write_capacity = var.env_table_write_capacity

  stream_enabled   = true
  stream_view_type = "NEW_AND_OLD_IMAGES"

  dynamic "replica" {
    for_each = var.replica_regions
    iterator = replica_region

    content {
      region_name = replica_region.value
    }
  }

  hash_key  = "deviceId"
  range_key = "epochS"
##...

Terraform's dynamic blocks repeat the named block once for each item assigned to the for_each meta argument. With the value for replica_regions that you set in terraform.tfvars, ["us-west-1", "ap-northeast-1"], the configuration above is equivalent to:

  replica {
    region_name = "us-west-1"
  }

  replica {
    region_name = "ap-northeast-1"
  }

Apply this change to configure global tables for your DynamoDB table. Respond to the confirmation prompt with a yes. It may take several minutes for Terraform to provision your global table replicas.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-TA894F2qiB5VWrS2

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_dynamodb_table.environment will be updated in-place
  ~ resource "aws_dynamodb_table" "environment" {
        id               = "environment_partially_clearly_polished_moth"
        name             = "environment_partially_clearly_polished_moth"
      ~ stream_enabled   = false -> true
      + stream_view_type = "NEW_AND_OLD_IMAGES"
        tags             = {}
        # (7 unchanged attributes hidden)

      + replica {
          + kms_key_arn = (known after apply)
          + region_name = "ap-northeast-1"
        }
      + replica {
          + kms_key_arn = (known after apply)
          + region_name = "us-west-1"
        }

        # (8 unchanged blocks hidden)
    }

Plan: 0 to add, 1 to change, 0 to destroy.


Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table.environment: Modifying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 10s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 20s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 30s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 40s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 50s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 1m0s elapsed]
aws_dynamodb_table.environment: Modifications complete after 1m5s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 0 added, 1 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Since AWS requires that you enable autoscaling before you enable global tables, you could not use Terraform to create a new dynamoDB table with both features enabled during a single apply action. Terraform would attempt to create your table and global tables before it created the autoscaling resources, and the AWS API would report an error:

│ Error: error initially creating DynamoDB Table (environment_partially_clearly_polished_moth) replicas:
error creating DynamoDB Table (environment_partially_clearly_polished_moth)
replica (ap-northeast-1): ValidationException: Table write capacity should
either be Pay-Per-Request or AutoScaled.

You can use dynamic blocks like the one in the example configuration above to work around this limitation. To provision a new table and then enable global tables using this configuration, you would:

  1. Apply the configuration with replica_regions set to the default value, [], to create the table and autoscaling configuration, but not configure any global table replicas.

  2. Update the replica_regions variable's value with the list of regions you want to replicate your table into.

  3. Apply the configuration again to enable global table replicas.

Alternatively, you could configure your table to use the PAY-PER-REQUEST billing (capacity) mode instead of using autoscaling. In that case, Terraform could create the table and enable global table replicas in the same apply step.

Refer to this issue for more information about this limitation.

Note

If you are using autoscaling, you must also enable autoscaling on any global secondary indexes for the table before you configure global tables.

Manage TTL

DynamoDB allows you to expire items with a Time To Live (TTL) attribute. When TTL is configured, AWS will automatically remove items whose TTL attribute contains a timestamp that has passed. This can help you manage your table's performance by removing stale items.

For example, your application may store data from your environmental sensors every 5 minutes, as well as hourly average data. Once the hourly average has been calculated, you may no longer need the data from every 5 minutes. Use TTL to remove that data automatically.

Add a ttl block to your DynamoDB table to enable TTL.

main.tf

resource "aws_dynamodb_table" "environment" {
  name         = random_pet.table_name.id
  billing_mode = "PROVISIONED"

  read_capacity  = var.env_table_read_capacity
  write_capacity = var.env_table_write_capacity

  ttl {
    enabled        = true
    attribute_name = "expiry"
  }

  stream_enabled   = true
  stream_view_type = "NEW_AND_OLD_IMAGES"
##...

Inside the ttl block, attribute_name specifies the DynamoDB table attribute which will contain the TTL as epoch time in seconds. After this time DynamoDB will delete the associated item. Do not include a separate attribute block for the TTL attribute in your Terraform configuration unless you will also use it in an index.

Apply this change to configure TTL for your DynamoDB table. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-pF3H7B9isccuHRGw

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Terraform used the selected providers to generate the following execution
plan. Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  # aws_dynamodb_table.environment will be updated in-place
  ~ resource "aws_dynamodb_table" "environment" {
        id               = "environment_partially_clearly_polished_moth"
        name             = "environment_partially_clearly_polished_moth"
        tags             = {}
        # (11 unchanged attributes hidden)

      ~ ttl {
          + attribute_name = "expiry"
          ~ enabled        = false -> true
        }
        # (9 unchanged blocks hidden)
    }

Plan: 0 to add, 1 to change, 0 to destroy.


Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table.environment: Modifying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Modifications complete after 4s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 0 added, 1 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Change the table class

In addition to the default Standard table class, DynamoDB supports Standard-Infrequent Access tables for rarely-accessed data. The Standard-Infrequent Access table class offers less expensive data storage, but more expensive data access and throughput.

Update your table's configuration to set the table class.

main.tf

resource "aws_dynamodb_table" "environment" {
  name         = random_pet.table_name.id
  billing_mode = "PROVISIONED"
  table_class  = "STANDARD_INFREQUENT_ACCESS"

  read_capacity  = var.env_table_read_capacity
  write_capacity = var.env_table_write_capacity
##...

Apply this change to configure your DynamoDB table's class. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/hashicorp-learn/learn-terraform-aws-dynamodb-scale/runs/run-9c6g8zpiwDHXsSzG

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Terraform will perform the following actions:

  # aws_dynamodb_table.environment will be updated in-place
  ~ resource "aws_dynamodb_table" "environment" {
        id               = "environment_partially_clearly_polished_moth"
        name             = "environment_partially_clearly_polished_moth"
      + table_class      = "STANDARD_INFREQUENT_ACCESS"
        tags             = {}
        # (11 unchanged attributes hidden)

        # (10 unchanged blocks hidden)
    }

Plan: 0 to add, 1 to change, 0 to destroy.


Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

Waiting for the apply to start...

aws_dynamodb_table.environment: Modifying... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 10s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 20s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 30s elapsed]
aws_dynamodb_table.environment: Still modifying... [id=environment_partially_clearly_polished_moth, 40s elapsed]
aws_dynamodb_table.environment: Modifications complete after 46s [id=environment_partially_clearly_polished_moth]

Apply complete! Resources: 0 added, 1 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Changing the table class does not require replacing the table. However, if you modify a table's class, the AWS API will not allow you to make any other changes to the table configuration in that same request.

Populate your table

The Terraform AWS provider includes an aws_dynamodb_table_item resource that you can use to manage items in your DynamoDB tables. While Terraform is not the appropriate tool to manage table items in most cases, you may wish to use this resource to populate static data or example items for a test environment.

First, add a variable definition to variables.tf to control whether or not Terraform will load example data.

variables.tf

variable "load_example_data" {
  description = "Flag: load example data into table items."
  type        = bool
  default     = true
}

Add the following configuration to main.tf to load example data from the data/example_environments.csv file into your table if the load_example_data variable is set to true.

main.tf

locals {
  example_data = csvdecode(file("data/example_environments.csv"))
}

resource "aws_dynamodb_table_item" "example" {
  for_each = var.load_example_data ? { for row in local.example_data : row.eventId => row } : {}

  table_name = aws_dynamodb_table.environment.name
  hash_key   = aws_dynamodb_table.environment.hash_key
  range_key  = aws_dynamodb_table.environment.range_key

  item = <<EOF
{
    "userId": {"S": "${each.value.userId}"},
    "deviceId": {"S": "${each.value.deviceId}"},
    "eventId": {"S": "${each.value.eventId}"},
    "geoLocation": {"S": "${each.value.geoLocation}"},
    "epochS": {"N": "${each.value.epochS}"},
    "expiry": {"N": "${each.value.expiry}"},
    "tempC": {"N": "${each.value.tempC}"},
    "humidityPct": {"N": "${each.value.humidityPct}"},
    "pressurePa": {"N": "${each.value.pressurePa}"}
}
EOF

  lifecycle {
    ignore_changes = [item]
  }
}

The aws_dynamodb_table_item.example block you added to main.tf uses a ternary operator to conditionally load the data based on the value of the load_example_data variable.

  for_each = var.load_example_data ? { for row in local.example_data : row.eventId => row } : {}

You can use this pattern to load example data in some environments, but not others. For example, you would set load_example_data to true when you provision your table in a test environment, and false when you provision your table in production. This way, your integration tests will work with a known set of data in the test environment, but that data will not exist in production.

Apply this change to load the example data into your DynamoDB table. Respond to the confirmation prompt with a yes.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-os5cg75SD8KKXYZp

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Terraform will perform the following actions:

  # aws_dynamodb_table_item.example["03568c98-c523-4faf-a3ef-575d2667d4b2"] will be created
  + resource "aws_dynamodb_table_item" "example" {

##...

Plan: 80 to add, 0 to change, 0 to destroy.


Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value: yes

aws_dynamodb_table_item.example["f41f52d4-8431-439b-9dc1-097280edfe9b"]: Creating...
aws_dynamodb_table_item.example["41f34be5-dafb-4fe3-856b-51354e327370"]: Creating...

##...

Apply complete! Resources: 80 added, 0 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Since global tables are enabled, AWS will automatically propagate this data to your replicas.

Query your data

Use the AWS CLI to query your DynamoDb table.

Note

If you are following this tutorial in HCP Terraform, first ensure that the AWS CLI is configured with the same credentials as your HCP Terraform workspace.

The TTL attribute, expiry, is set to a time in the past for 10 of the items in your table, so AWS will automatically delete them. AWS usually removes expired items very quickly. However, occasionally this process can take 48 hours or more. Because of this, the data you get back from the following queries may be different from the examples below.

Query your table for events for a specific device ID, b6c772c6-d621-46ff-86c6-7c662de62375.

$ aws dynamodb query --table-name $(terraform output -raw environment_table_name) \
    --region us-east-1 \
    --key-condition-expression "deviceId = :device" \
    --expression-attribute-values '{":device": {"S": "b6c772c6-d621-46ff-86c6-7c662de62375"}}' \
    --no-cli-pager

The AWS CLI This query will return either 10 results, or 8 results if AWS has already removed expired items.

{
    "Items": [
        {
            "geoLocation": {
                "S": "Earth-US-VT-St. Albans"
            },
            "eventId": {
                "S": "93dbe64e-9e49-4fcb-a4fe-f2d6a1c2dabf"
            },
            "pressurePa": {
                "N": "104750"
            },
            "deviceId": {
                "S": "b6c772c6-d621-46ff-86c6-7c662de62375"
            },
            "userId": {
                "S": "49025614-5f66-4bd8-ba79-fd9dc9f75e6f"
            },
            "epochS": {
                "N": "1649257045"
            },
            "tempC": {
                "N": "99.56"

##...

    ],
    "Count": 8,
    "ScannedCount": 8,
    "ConsumedCapacity": null
}

Query events by device ID, within a time range, using the us-west-1 global table replica.

$ aws dynamodb query --table-name $(terraform output -raw environment_table_name) \
    --region us-west-1 \
    --key-condition-expression "deviceId = :device and (epochS between :start and :end)" \
    --expression-attribute-values '{":device": {"S": "b6c772c6-d621-46ff-86c6-7c662de62375"}, ":start": {"N": "1649257000"}, ":end": {"N": "1649260000"}}' \
    --no-cli-pager

This query will return either 5 results, or 4 results if AWS has already removed expired items.

{
    "Items": [
        {
            "geoLocation": {
                "S": "Earth-US-VT-St. Albans"
            },

##...

    ],
    "Count": 4,
    "ScannedCount": 4,
    "ConsumedCapacity": null
}

Query events by geoLocation.

$ aws dynamodb query --table-name $(terraform output -raw environment_table_name) \
    --region us-east-1 \
    --index-name by_geoLocation \
    --key-condition-expression "geoLocation = :loc" \
    --expression-attribute-values '{":loc": {"S": "Earth-US-CA-Berkely"}}' \
    --no-cli-pager

This query will return either 20 results, or 18 results if AWS has already removed expired items.

{
    "Items": [
        {
            "geoLocation": {
                "S": "Earth-US-CA-Berkely"
            },

##...
    ],
    "Count": 18,
    "ScannedCount": 18,
    "ConsumedCapacity": null
}

Observe the effects of TTL

Your configuration manages table items with the aws_dynamodb_table_item.example block. Terraform will verify that the items exist whenever you plan or apply changes. Since the TTL is set for several of the items, AWS will remove them automatically. After it does so, Terraform will notice that they are missing and attempt to recreate them whenever you apply your configuration. Depending on your use case, you may want to avoid this behavior by not setting the TTL attribute for table items managed by Terraform.

Apply your configuration now.

$ terraform apply
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-KBmaxecSChwb9HGQ

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Plan: 10 to add, 0 to change, 0 to destroy.

Do you want to perform these actions in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will perform the actions described above.
  Only 'yes' will be accepted to approve.

  Enter a value:

If AWS has removed the items with the TTL attribute, expiry, set, Terraform will prompt you to add the 10 expired table items. Otherwise, it will just plan the changes to your global table's read and write capacity, as described above.

Respond to the confirmation prompt with a yes, and Terraform will add the expired items. AWS will once again remove them because of the TTL configured in the expiry attribute.

  Enter a value: yes

aws_dynamodb_table_item.example["20c49430-08db-41ae-89b9-0247e47d83c5"]: Creating...
aws_dynamodb_table_item.example["32fdec0f-ceeb-4ad7-a491-595b50fcb8c3"]: Creating...

##...

Apply complete! Resources: 10 added, 0 changed, 0 destroyed.

Outputs:

environment_table_arn = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth"
environment_table_name = "environment_partially_clearly_polished_moth"

Clean up infrastructure

Remove your table and related resources. Respond to the confirmation prompt with a yes.

$ terraform destroy
Running apply in HCP Terraform. Output will stream here. Pressing Ctrl-C
will cancel the remote apply if it's still pending. If the apply started it
will stop streaming the logs, but will not stop the apply running remotely.

Preparing the remote apply...

To view this run in a browser, visit:
https://app.terraform.io/app/YOUR-ORG/learn-terraform-aws-dynamodb-scale/runs/run-F4rQ8gsCtUPRU8WZ

Waiting for the plan to start...

Terraform v1.2.4
on linux_amd64
Initializing plugins and modules...
random_pet.table_name: Refreshing state... [id=environment_partially_clearly_polished_moth]
aws_dynamodb_table.environment: Refreshing state... [id=environment_partially_clearly_polished_moth]

##...

Plan: 0 to add, 0 to change, 80 to destroy.

Changes to Outputs:
  - environment_table_arn  = "arn:aws:dynamodb:us-east-1:561656980159:table/environment_partially_clearly_polished_moth" -> null
  - environment_table_name = "environment_partially_clearly_polished_moth" -> null

Do you really want to destroy all resources in workspace "learn-terraform-aws-dynamodb-scale"?
  Terraform will destroy all your managed infrastructure, as shown above.
  There is no undo. Only 'yes' will be accepted to confirm.

  Enter a value: yes

aws_dynamodb_table_item.example["d6523155-cf06-4d5d-b3dc-c5007fbee732"]: Destroying... [id=environment_partially_clearly_polished_moth|deviceId||b6c772c6-d621-46ff-86c6-7c662de62375||epochS|||1649258473]
aws_dynamodb_table_item.example["e5acbef5-5674-4f61-8d92-8d7e8055e6b8"]: Destroying... [id=environment_partially_clearly_polished_moth|deviceId||05302738-ceba-4218-916a-82503d2bee91||epochS|||1649258808]

##...

random_pet.table_name: Destruction complete after 0s

Apply complete! Resources: 0 added, 0 changed, 80 destroyed.

Note

Destroying your DynamoDB table automatically destroys any items in the table, whether they were created by Terraform or not. Terraform will explicitly destroy the items it manages, and AWS will automatically remove any others when Terraform destroys the table.

If you used HCP Terraform for this tutorial, after destroying your resources, delete the learn-terraform-aws-dynamodb-scale workspace from your HCP Terraform organization.

Next steps

In this tutorial, you provisioned a DynamoDB table and then configured features related to managing your table's performance and capacity. You also provisioned items in your table, and ran several queries for the example data. Review the following resources to learn more about managing AWS DynamoDB with Terraform.

Previous

Vercel preview environments

 Next Collection

Azure Services