Installation - Cloud-Based Nodes¶
Anaconda Adam can be used to create cloud-based instances and install Anaconda platform components on the nodes.
Currently, Amazon EC2 is fully supported. To use Adam with other cloud providers or instances that already exist on Amazon EC2, you can provision the nodes following the primary installation instructions.
Refer to the Requirements page for more information about system requirements, including networking and security requirements.
1. Download Adam¶
Download the installer to the host machine where you want to manage the cloud-based Anaconda platform:
$ wget https://anaconda.org/anaconda-adam/adam-installer/4.2.3/download/adam-installer-4.2.3-Linux-x86_64.sh
or:
$ curl -LO https://anaconda.org/anaconda-adam/adam-installer/4.2.3/download/adam-installer-4.2.3-Linux-x86_64.sh
2. Run the Installer¶
Run the installer, and specify the destination directory (e.g., ~/adam
):
$ bash adam-installer-4.2.3-Linux-x86_64.sh -b -p ~/adam
Add the adam
command to your path by modifying ~/.bashrc
:
$ echo -e '\n# Anaconda Adam\nexport PATH=~/adam/bin:$PATH' >> ~/.bashrc
Source your ~/.bashrc
file to activate the changes:
$ source ~/.bashrc
Run the adam
command to verify the installation and to create the configuration directories:
$ adam
3. License File (Optional)¶
If you have an Anaconda Workgroup, Anaconda Enterprise, or other Anaconda
license file, you can copy it to the ~/adam/licenses
directory. Anaconda
Adam will search this directory for license files named license*.txt
, e.g.,
~/adam/licenses/license_bundle_20160524191753.txt
.
TIP: In addition to ~/adam/licenses/
, Adam checks for license files in
~/.continuum/
and the other directories shown by the command
~/adam/bin/conda info --license
, so you may place the license file in any of
these directories.
4. Configure AWS credentials¶
Configure your AWS credentials including your AWS Access Key ID, AWS Secret
Access Key, and (optionally) your default region in a file named
~/.aws/credentials
. An example credentials file is shown below:
[default]
aws_access_key_id = YOUR_ACCESS_KEY
aws_secret_access_key = YOUR_SECRET_KEY
region = us-east-1
Refer to the boto documentation for more information on configuring your AWS credentials.
5. Create Cluster¶
Create your cluster on Amazon EC2 using:
$ adam ec2 -n my-cluster up
The keyname and keypair are required arguments. You can optionally specify the AMI, number of nodes, instance types, etc.
$ adam ec2 -n my-cluster up --help
Usage: adam ec2 up [OPTIONS]
Options:
--keyname TEXT Keyname on EC2 console [required]
--keypair PATH Path to the keypair that matches the keyname
[required]
--vpc-id TEXT EC2 VPC ID
--subnet-id TEXT EC2 Subnet ID on the VPC
--region-name TEXT AWS region [default: us-east-1]
--ami TEXT EC2 AMI [default: ami-d05e75b8]
--username TEXT User to SSH to the AMI [default: ubuntu]
--type TEXT EC2 Instance Type [default: m4.xlarge]
--count INTEGER Number of nodes [default: 4]
--security-group TEXT Security Group Name [default: adam-default]
--volume-type TEXT Root volume type [default: gp2]
--volume-size INTEGER Root volume size (GB) [default: 500]
-t, --tag TEXT Extra tags to add to the instances
--check-ami / --no-check-ami Whether or not to check the AMI [default:
True]
--provision / --no-provision Provision salt on the nodes [default: True]
-y, --yes Answers yes to questions
-h, --help Show this message and exit.
6. Install Platform Components¶
After creating a cluster, you can install platform components using
adam <platform-component> install
:
$ adam repository -n my-cluster install
$ adam enterprise-notebooks -n my-cluster install
$ adam mosaic -n my-cluster install
$ adam accelerate -n my-cluster install
$ adam scale -n my-cluster cluster install
$ adam scale -n my-cluster dask install
7. Access Platform Components¶
After the platform component installations complete, you can open the platform
component interfaces in your browser using adam <platform-component> open
:
$ adam repository -n my-cluster open
$ adam enterprise-notebooks -n my-cluster open
$ adam mosaic -n my-cluster open
Refer to the Getting Started page for more information about using the Anaconda platform components.