DAX

Overview

CloudMock emulates Amazon DynamoDB Accelerator (DAX), providing a fully in-memory mock of the DAX control plane API. DAX is an in-memory cache for DynamoDB that delivers up to 10x performance improvement. CloudMock supports cluster management, subnet groups, parameter groups, and tagging.

Supported Operations

Operation	Status	Notes
CreateCluster	Supported	Creates a DAX cluster with proper ARN (arn:aws:dax:{region}:{account}:cache/{name})
DescribeClusters	Supported	Lists all clusters or filters by name
UpdateCluster	Supported	Updates description, maintenance window, notification topic
DeleteCluster	Supported	Deletes a cluster (returns cluster in deleting state)
IncreaseReplicationFactor	Supported	Adds nodes to an existing cluster
DecreaseReplicationFactor	Supported	Removes nodes from an existing cluster
CreateSubnetGroup	Supported	Creates a subnet group
DescribeSubnetGroups	Supported	Lists subnet groups
DeleteSubnetGroup	Supported	Deletes a subnet group
CreateParameterGroup	Supported	Creates a parameter group
DescribeParameterGroups	Supported	Lists parameter groups
UpdateParameterGroup	Supported	Updates parameters in a parameter group
DeleteParameterGroup	Supported	Deletes a parameter group
DescribeParameters	Supported	Returns parameters for a given group
DescribeDefaultParameters	Supported	Returns default DAX parameters
TagResource	Supported	Adds tags to a cluster by ARN
UntagResource	Supported	Removes tags from a cluster
ListTags	Supported	Lists tags for a cluster

Cluster States

DAX clusters progress through the following states:

State	Description
creating	Cluster is being provisioned
available	Cluster is ready for use
modifying	Cluster configuration is being updated
deleting	Cluster is being deleted

In CloudMock, clusters transition from creating to available nearly instantly (lifecycle machine with delays disabled in test mode).

Quick Start

Node.js

import { DAXClient, CreateClusterCommand, DescribeClustersCommand } from '@aws-sdk/client-dax';

const client = new DAXClient({
  endpoint: 'http://localhost:4566',
  region: 'us-east-1',
  credentials: { accessKeyId: 'test', secretAccessKey: 'test' },
});

// Create a cluster
const { Cluster } = await client.send(new CreateClusterCommand({
  ClusterName: 'my-dax-cluster',
  NodeType: 'dax.r4.large',
  ReplicationFactor: 3,
  IamRoleArn: 'arn:aws:iam::123456789012:role/AmazonDAXFullAccess',
  SubnetGroupName: 'my-subnet-group',
}));
console.log(Cluster.ClusterArn);
// arn:aws:dax:us-east-1:123456789012:cache/my-dax-cluster

// List clusters
const { Clusters } = await client.send(new DescribeClustersCommand({}));
console.log(Clusters.length);

Python

import boto3

client = boto3.client('dax',
    endpoint_url='http://localhost:4566',
    region_name='us-east-1',
    aws_access_key_id='test',
    aws_secret_access_key='test')

# Create a cluster
response = client.create_cluster(
    ClusterName='my-dax-cluster',
    NodeType='dax.r4.large',
    ReplicationFactor=3,
    IamRoleArn='arn:aws:iam::123456789012:role/AmazonDAXFullAccess',
    SubnetGroupName='my-subnet-group')
print(response['Cluster']['ClusterArn'])

# Create a parameter group
pg = client.create_parameter_group(
    ParameterGroupName='my-params',
    Description='Custom DAX parameters')

# Update a parameter
client.update_parameter_group(
    ParameterGroupName='my-params',
    ParameterNameValues=[
        {'ParameterName': 'query-ttl-millis', 'ParameterValue': '600000'},
        {'ParameterName': 'record-ttl-millis', 'ParameterValue': '600000'},
    ])

Configuration

cloudmock.yml

services:
  dax:
    enabled: true

ARN Format

DAX cluster ARNs follow this format:

arn:aws:dax:{region}:{account-id}:cache/{cluster-name}

Default Parameters

CloudMock provides the following default DAX parameters:

Parameter	Default Value	Description
query-ttl-millis	300000	TTL for cached query results (5 minutes)
record-ttl-millis	300000	TTL for cached item records (5 minutes)

Known Differences from AWS

• Cluster endpoints are synthetic DNS names and do not accept real DAX protocol connections
• Node types are not validated against available DAX node type catalog
• Availability zone placement is simulated
• Cluster encryption (SSE) is tracked but not enforced
• No VPC subnet validation is performed