diff --git a/cloud/amazon/redshift.py b/cloud/amazon/redshift.py
new file mode 100644
index 00000000000..b6c6fdd35d5
--- /dev/null
+++ b/cloud/amazon/redshift.py
@@ -0,0 +1,502 @@
+#!/usr/bin/python
+
+# Copyright 2014 Jens Carl, Hothead Games Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+DOCUMENTATION = '''
+---
+author: 
+  - "Jens Carl (@j-carl), Hothead Games Inc."
+module: redshift
+version_added: "2.1"
+short_description: create, delete, or modify an Amazon Redshift instance
+description:
+  - Creates, deletes, or modifies amazon Redshift cluster instances.
+options:
+  command:
+    description:
+      - Specifies the action to take.
+    required: true
+    choices: [ 'create', 'facts', 'delete', 'modify' ]
+  identifier:
+    description:
+      - Redshift cluster identifier.
+    required: true
+  node_type:
+    description:
+      - The node type of the cluster. Must be specified when command=create.
+    required: false
+    choices: ['dw1.xlarge', 'dw1.8xlarge', 'dw2.large', 'dw2.8xlarge', ]
+  username:
+    description:
+      - Master database username. Used only when command=create.
+    required: false
+  password:
+    description:
+      - Master database password. Used only when command=create.
+    required: false
+  cluster_type:
+    description:
+      - The type of cluster.
+    required: false
+    choices: ['multi-node', 'single-node' ]
+    default: 'single-node'
+  db_name:
+    description:
+      - Name of the database.
+    required: false
+    default: null
+  availability_zone:
+    description:
+      - availability zone in which to launch cluster
+    required: false
+    aliases: ['zone', 'aws_zone']
+  number_of_nodes:
+    description:
+      - Number of nodes. Only used when cluster_type=multi-node.
+    required: false
+    default: null
+  cluster_subnet_group_name:
+    description:
+      - which subnet to place the cluster
+    required: false
+    aliases: ['subnet']
+  cluster_security_groups:
+    description:
+      - in which security group the cluster belongs
+    required: false
+    default: null
+    aliases: ['security_groups']
+  vpc_security_group_ids:
+    description:
+      - VPC security group
+    required: false
+    aliases: ['vpc_security_groups']
+    default: null
+  preferred_maintenance_window:
+    description:
+      - maintenance window
+    required: false
+    aliases: ['maintance_window', 'maint_window']
+    default: null
+  cluster_parameter_group_name:
+    description:
+      - name of the cluster parameter group
+    required: false
+    aliases: ['param_group_name']
+    default: null
+  automated_snapshot_retention_period:
+    description:
+      - period when the snapshot take place
+    required: false
+    aliases: ['retention_period']
+    default: null
+  port:
+    description:
+      - which port the cluster is listining
+    required: false
+    default: null
+  cluster_version:
+    description:
+      - which version the cluster should have
+    required: false
+    aliases: ['version']
+    choices: ['1.0']
+    default: null
+  allow_version_upgrade:
+    description:
+      - flag to determinate if upgrade of version is possible
+    required: false
+    aliases: ['version_upgrade']
+    default: null
+  number_of_nodes:
+    description:
+      - number of the nodes the cluster should run
+    required: false
+    default: null
+  publicly_accessible:
+    description:
+      - if the cluster is accessible publicly or not
+    required: false
+    default: null
+  encrypted:
+    description:
+      -  if the cluster is encrypted or not
+    required: false
+    default: null
+  elastic_ip:
+    description:
+      - if the cluster has an elastic IP or not
+    required: false
+    default: null
+  new_cluster_identifier:
+    description:
+      - Only used when command=modify.
+    required: false
+    aliases: ['new_identifier']
+    default: null
+  wait:
+    description:
+      - When command=create, modify or restore then wait for the database to enter the 'available' state. When command=delete wait for the database to be terminated.
+    required: false
+    default: "no"
+    choices: [ "yes", "no" ]
+  wait_timeout:
+    description:
+      - how long before wait gives up, in seconds
+    default: 300
+requirements: [ 'boto' ]
+extends_documentation_fragment: aws
+'''
+
+EXAMPLES = '''
+# Basic cluster provisioning example
+- redshift: >
+    command=create
+    node_type=dw1.xlarge
+    identifier=new_cluster
+    username=cluster_admin
+    password=1nsecure
+'''
+
+RETURN = '''
+cluster:
+    description: dictionary containing all the cluster information
+    returned: success
+    type: dictionary
+    contains:
+        identifier:
+            description: Id of the cluster.
+            returned: success
+            type: string
+            sample: "new_redshift_cluster"
+        create_time:
+            description: Time of the cluster creation as timestamp.
+            returned: success
+            type: float
+            sample: 1430158536.308
+        status:
+            description: Stutus of the cluster.
+            returned: success
+            type: string
+            sample: "available"
+        db_name:
+            description: Name of the database.
+            returned: success
+            type: string
+            sample: "new_db_name"
+        availability_zone:
+            description: Amazon availability zone where the cluster is located.
+            returned: success
+            type: string
+            sample: "us-east-1b"
+        maintenance_window:
+            description: Time frame when maintenance/upgrade are done.
+            returned: success
+            type: string
+            sample: "sun:09:30-sun:10:00"
+        private_ip_address:
+            description: Private IP address of the main node.
+            returned: success
+            type: string
+            sample: "10.10.10.10"
+        public_ip_address:
+            description: Public IP address of the main node.
+            returned: success
+            type: string
+            sample: "0.0.0.0"
+        port:
+            description: Port of the cluster.
+            returned: success
+            type: int
+            sample: 5439
+        url:
+            description: FQDN of the main cluster node.
+            returned: success
+            type: string
+            sample: "new-redshift_cluster.jfkdjfdkj.us-east-1.redshift.amazonaws.com"
+'''
+
+import time
+
+try:
+    import boto
+    from boto import redshift
+    HAS_BOTO = True
+except ImportError:
+    HAS_BOTO = False
+
+
+def _collect_facts(resource):
+    """Transfrom cluster information to dict."""
+    facts = {
+        'identifier'        : resource['ClusterIdentifier'],
+        'create_time'       : resource['ClusterCreateTime'],
+        'status'            : resource['ClusterStatus'],
+        'username'          : resource['MasterUsername'],
+        'db_name'           : resource['DBName'],
+        'availability_zone' : resource['AvailabilityZone'],
+        'maintenance_window': resource['PreferredMaintenanceWindow'],
+    }
+
+    for node in resource['ClusterNodes']:
+        if node['NodeRole'] in ('SHARED', 'LEADER'):
+            facts['private_ip_address'] = node['PrivateIPAddress']
+            break
+
+    return facts
+
+
+def create_cluster(module, redshift):
+    """
+    Create a new cluster
+
+    module: AnsibleModule object
+    redshift: authenticated redshift connection object
+
+    Returns:
+    """
+
+    identifier   = module.params.get('identifier')
+    node_type    = module.params.get('node_type')
+    username     = module.params.get('username')
+    password     = module.params.get('password')
+    wait         = module.params.get('wait')
+    wait_timeout = module.params.get('wait_timeout')
+
+    changed = True
+    # Package up the optional parameters
+    params = {}
+    for p in ('db_name', 'cluster_type', 'cluster_security_groups',
+              'vpc_security_group_ids', 'cluster_subnet_group_name',
+              'availability_zone', 'preferred_maintenance_window',
+              'cluster_parameter_group_name',
+              'automated_snapshot_retention_period', 'port',
+              'cluster_version', 'allow_version_upgrade',
+              'number_of_nodes', 'publicly_accessible',
+              'encrypted', 'elastic_ip'):
+        if module.params.get( p ):
+            params[ p ] = module.params.get( p )
+
+    try:
+        redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+        changed = False
+    except boto.exception.JSONResponseError, e:
+        try:
+            redshift.create_cluster(identifier, node_type, username, password, **params)
+        except boto.exception.JSONResponseError, e:
+            module.fail_json(msg=str(e))
+
+    try:
+        resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    if wait:
+        try:
+            wait_timeout = time.time() + wait_timeout
+            time.sleep(5)
+
+            while wait_timeout > time.time() and resource['ClusterStatus'] != 'available':
+                time.sleep(5)
+                if wait_timeout <= time.time():
+                    module.fail_json(msg = "Timeout waiting for resource %s" % resource.id)
+
+                resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+
+        except boto.exception.JSONResponseError, e:
+            module.fail_json(msg=str(e))
+
+    return(changed, _collect_facts(resource))
+
+
+def describe_cluster(module, redshift):
+    """
+    Collect data about the cluster.
+
+    module: Ansible module object
+    redshift: authenticated redshift connection object
+    """
+    identifier = module.params.get('identifier')
+
+    try:
+        resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    return(True, _collect_facts(resource))
+
+
+def delete_cluster(module, redshift):
+    """
+    Delete a cluster.
+
+    module: Ansible module object
+    redshift: authenticated redshift connection object
+    """
+
+    identifier   = module.params.get('identifier')
+    wait         = module.params.get('wait')
+    wait_timeout = module.params.get('wait_timeout')
+
+    try:
+        redshift.delete_custer( identifier )
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    if wait:
+        try:
+            wait_timeout = time.time() + wait_timeout
+            resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+
+            while wait_timeout > time.time() and resource['ClusterStatus'] != 'deleting':
+                time.sleep(5)
+                if wait_timeout <= time.time():
+                    module.fail_json(msg = "Timeout waiting for resource %s" % resource.id)
+
+                resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+
+        except boto.exception.JSONResponseError, e:
+            module.fail_json(msg=str(e))
+
+    return(True, {})
+
+
+def modify_cluster(module, redshift):
+    """
+    Modify an existing cluster.
+
+    module: Ansible module object
+    redshift: authenticated redshift connection object
+    """
+
+    identifier   = module.params.get('identifier')
+    wait         = module.params.get('wait')
+    wait_timeout = module.params.get('wait_timeout')
+
+    # Package up the optional parameters
+    params = {}
+    for p in ('cluster_type', 'cluster_security_groups',
+              'vpc_security_group_ids', 'cluster_subnet_group_name',
+              'availability_zone', 'preferred_maintenance_window',
+              'cluster_parameter_group_name',
+              'automated_snapshot_retention_period', 'port', 'cluster_version',
+              'allow_version_upgrade', 'number_of_nodes', 'new_cluster_identifier'):
+        if module.params.get(p):
+            params[p] = module.params.get(p)
+
+    try:
+        redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+        changed = False
+    except boto.exception.JSONResponseError, e:
+        try:
+            redshift.modify_cluster(identifier, **params)
+        except boto.exception.JSONResponseError, e:
+            module.fail_json(msg=str(e))
+
+    try:
+        resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    if wait:
+        try:
+            wait_timeout = time.time() + wait_timeout
+            time.sleep(5)
+
+            while wait_timeout > time.time() and resource['ClusterStatus'] != 'available':
+                time.sleep(5)
+                if wait_timeout <= time.time():
+                    module.fail_json(msg = "Timeout waiting for resource %s" % resource.id)
+
+                resource = redshift.describe_clusters(identifier)['DescribeClustersResponse']['DescribeClustersResult']['Clusters'][0]
+
+        except boto.exception.JSONResponseError, e:
+            # https://github.com/boto/boto/issues/2776 is fixed.
+            module.fail_json(msg=str(e))
+
+    return(True, _collect_facts(resource))
+
+
+def main():
+    argument_spec = ec2_argument_spec()
+    argument_spec.update(dict(
+            command                             = dict(choices=['create', 'facts', 'delete', 'modify'], required=True),
+            identifier                          = dict(required=True),
+            node_type                           = dict(choices=['dw1.xlarge', 'dw1.8xlarge', 'dw2.large', 'dw2.8xlarge', ], required=False),
+            username                            = dict(required=False),
+            password                            = dict(no_log=True, required=False),
+            db_name                             = dict(require=False),
+            cluster_type                        = dict(choices=['multi-node', 'single-node', ], default='single-node'),
+            cluster_security_groups             = dict(aliases=['security_groups'], type='list'),
+            vpc_security_group_ids              = dict(aliases=['vpc_security_groups'], type='list'),
+            cluster_subnet_group_name           = dict(aliases=['subnet']),
+            availability_zone                   = dict(aliases=['aws_zone', 'zone']),
+            preferred_maintenance_window        = dict(aliases=['maintance_window', 'maint_window']),
+            cluster_parameter_group_name        = dict(aliases=['param_group_name']),
+            automated_snapshot_retention_period = dict(aliases=['retention_period']),
+            port                                = dict(type='int'),
+            cluster_version                     = dict(aliases=['version'], choices=['1.0']),
+            allow_version_upgrade               = dict(aliases=['version_upgrade'], type='bool'),
+            number_of_nodes                     = dict(type='int'),
+            publicly_accessible                 = dict(type='bool'),
+            encrypted                           = dict(type='bool'),
+            elastic_ip                          = dict(required=False),
+            new_cluster_identifier              = dict(aliases=['new_identifier']),
+            wait                                = dict(type='bool', default=False),
+            wait_timeout                        = dict(default=300),
+        )
+    )
+
+    module = AnsibleModule(
+        argument_spec=argument_spec,
+    )
+
+    if not HAS_BOTO:
+        module.fail_json(msg='boto v2.9.0+ required for this module')
+
+    command = module.params.get('command')
+
+    region, ec2_url, aws_connect_params = get_aws_connection_info(module)
+    if not region:
+        module.fail_json(msg=str("region not specified and unable to determine region from EC2_REGION."))
+
+    # connect to the rds endpoint
+    try:
+        conn = connect_to_aws(boto.redshift, region, **aws_connect_params)
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    changed = True
+    if command == 'create':
+        (changed, cluster) = create_cluster(module, conn)
+
+    elif command == 'facts':
+        (changed, cluster) = describe_cluster(module, conn)
+
+    elif command == 'delete':
+        (changed, cluster) = delete_cluster(module, conn)
+
+    elif command == 'modify':
+        (changed, cluster) = modify_cluster(module, conn)
+
+    module.exit_json(changed=changed, cluster=cluster)
+
+# import module snippets
+from ansible.module_utils.basic import *
+from ansible.module_utils.ec2 import *
+
+if __name__ == '__main__':
+    main()
diff --git a/cloud/amazon/redshift_subnet_group.py b/cloud/amazon/redshift_subnet_group.py
new file mode 100644
index 00000000000..acd3330c1f2
--- /dev/null
+++ b/cloud/amazon/redshift_subnet_group.py
@@ -0,0 +1,182 @@
+#!/usr/bin/python
+
+# Copyright 2014 Jens Carl, Hothead Games Inc.
+#
+# This program is free software: you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+# the Free Software Foundation, either version 3 of the License, or
+# (at your option) any later version.
+#
+# This program is distributed in the hope that it will be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
+#
+# You should have received a copy of the GNU General Public License
+# along with this program. If not, see <http://www.gnu.org/licenses/>.
+
+DOCUMENTATION = '''
+---
+author: 
+  - "Jens Carl (@j-carl), Hothead Games Inc."
+module: redshift_subnet_group
+version_added: "2.1"
+short_description: mange Redshift cluster subnet groups
+description:
+  - Create, modifies, and deletes Redshift cluster subnet groups.
+options:
+  state:
+    description:
+      - Specifies whether the subnet should be present or absent.
+    default: 'present'
+    choices: ['present', 'absent' ]
+  group_name:
+    description:
+      - Cluster subnet group name.
+    required: true
+    aliases: ['name']
+  group_description:
+    description:
+      - Database subnet group description.
+    required: false
+    default: null
+    aliases: ['description']
+  group_subnets:
+    description:
+      - List of subnet IDs that make up the cluster subnet group.
+    required: false
+    default: null
+    aliases: ['subnets']
+requirements: [ 'boto' ]
+extends_documentation_fragment: aws
+'''
+
+EXAMPLES = '''
+# Create a Redshift subnet group
+- local_action:
+    module: redshift_subnet_group
+    state: present
+    group_name: redshift-subnet
+    group_description: Redshift subnet
+    group_subnets:
+        - 'subnet-aaaaa'
+        - 'subnet-bbbbb'
+
+# Remove subnet group
+redshift_subnet_group: >
+    state: absent
+    group_name: redshift-subnet
+'''
+
+RETURN = '''
+group:
+    description: dictionary containing all Redshift subnet group information
+    returned: success
+    type: dictionary
+    contains:
+        name:
+            description: name of the Redshift subnet group
+            returned: success
+            type: string
+            sample: "redshift_subnet_group_name"
+        vpc_id:
+            description: Id of the VPC where the subnet is located
+            returned: success
+            type: stering
+            sample: "vpc-aabb1122"
+'''
+
+try:
+    import boto
+    import boto.redshift
+    HAS_BOTO = True
+except ImportError:
+    HAS_BOTO = False
+
+
+def main():
+    argument_spec = ec2_argument_spec()
+    argument_spec.update(dict(
+        state=dict(required=True, choices=['present', 'absent']),
+        group_name=dict(required=True, aliases=['name']),
+        group_description=dict(required=False, aliases=['description']),
+        group_subnets=dict(required=False, aliases=['subnets'], type='list'),
+    ))
+    module = AnsibleModule(argument_spec=argument_spec)
+
+    if not HAS_BOTO:
+        module.fail_json(msg='boto v2.9.0+ required for this module')
+
+    state = module.params.get('state')
+    group_name = module.params.get('group_name')
+    group_description = module.params.get('group_description')
+    group_subnets = module.params.get('group_subnets')
+
+    if state == 'present':
+        for required in ('group_name', 'group_description', 'group_subnets'):
+            if not module.params.get(required):
+                module.fail_json(msg=str("parameter %s required for state='present'" % required))
+    else:
+        for not_allowed in ('group_description', 'group_subnets'):
+            if module.params.get(not_allowed):
+                module.fail_json(msg=str("parameter %s not allowed for state='absent'" % not_allowed))
+
+    region, ec2_url, aws_connect_params = get_aws_connection_info(module)
+    if not region:
+        module.fail_json(msg=str("region not specified and unable to determine region from EC2_REGION."))
+
+    # Connect to the Redshift endpoint.
+    try:
+        conn = connect_to_aws(boto.redshift, region, **aws_connect_params)
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    try:
+        changed = False
+        exists = False
+        group = None
+
+        try:
+            matching_groups = conn.describe_cluster_subnet_groups(group_name, max_records=100)
+            exists = len(matching_groups) > 0
+        except boto.exception.JSONResponseError, e:
+            if e.body['Error']['Code'] != 'ClusterSubnetGroupNotFoundFault':
+            #if e.code != 'ClusterSubnetGroupNotFoundFault':
+                module.fail_json(msg=str(e))
+
+        if state == 'absent':
+            if exists:
+                conn.delete_cluster_subnet_group(group_name)
+                changed = True
+
+        else:
+            if not exists:
+                new_group = conn.create_cluster_subnet_group(group_name, group_description, group_subnets)
+                group = {
+                    'name': new_group['CreateClusterSubnetGroupResponse']['CreateClusterSubnetGroupResult']
+                            ['ClusterSubnetGroup']['ClusterSubnetGroupName'],
+                    'vpc_id': new_group['CreateClusterSubnetGroupResponse']['CreateClusterSubnetGroupResult']
+                              ['ClusterSubnetGroup']['VpcId'],
+                }
+            else:
+                changed_group = conn.modify_cluster_subnet_group(group_name, group_subnets, description=group_description)
+                group = {
+                    'name': changed_group['ModifyClusterSubnetGroupResponse']['ModifyClusterSubnetGroupResult']
+                            ['ClusterSubnetGroup']['ClusterSubnetGroupName'],
+                    'vpc_id': changed_group['ModifyClusterSubnetGroupResponse']['ModifyClusterSubnetGroupResult']
+                              ['ClusterSubnetGroup']['VpcId'],
+                }
+
+            changed = True
+
+    except boto.exception.JSONResponseError, e:
+        module.fail_json(msg=str(e))
+
+    module.exit_json(changed=changed, group=group)
+
+# import module snippets
+from ansible.module_utils.basic import *
+from ansible.module_utils.ec2 import *
+
+if __name__ == '__main__':
+    main()