Home
Configure Amazon Web Service Cloud Scaling
Create the necessary components in AWS that are required for cloud scaling and then add your credential to NavOps before creating a cloud environment.
Multi-Availability Zone Management on AWS
Configure the cloud environment so that cloud nodes can be deployed in several Availability Zones.

Welcome
What's New
New and updated features in Altair NavOps 2025.1.0.
Get Started
Cloud cost tracking, Automation based scalability, and Cloud migration for HPC.
Cloud Scaling Overview
An introduction to cloud scaling, a typical scaling setup and an overview of steps to set up cloud scaling.
Configure Amazon Web Service Cloud Scaling
Create the necessary components in AWS that are required for cloud scaling and then add your credential to NavOps before creating a cloud environment.
- Create an Amazon Web Service User Account
  Create an Amazon Web Service (AWS) user account so that you can later add a cloud credential in NavOps.
- Add an Amazon Web Service Credential
  Add an existing AWS user account to NavOps so that you can create a cloud environment.
- Create a Virtual Private Cloud Network
  Create a Virtual Private Cloud (VPC) so that you can later create a virtual machine for cloud scaling.
- Create an Internet Gateway
  Create an internet gateway and attach it to the scaling VPC.
- Create and Edit the VPC Route Table
  Add a rule to the VPC route table that allows all internet access and associate the route table to the scaling subnet.
- Add Inbound Rules to the VPC Security Group
  Add inbound rules to the VPC security group so that a connection can be established with an AWS VM using SSH or RDP.
- Create a Virtual Machine
  Create a virtual machine in AWS Elastic Compute Cloud (EC2).
- Install Accelerator on the VM
  Install Accelerator on the Linux virtual machine and configure the VM per your site's environment.
- Install PBS MoM on the VM
  Install the PBS Professional MoM on the virtual machine and configure the VM per your site's environment.
- Install Grid Engine Execd on the VM
  Install Grid Engine Execd on the Linux virtual machine and configure the VM per your site's environment.
- Install Slurm on the VM
  Install Slurm on the Linux virtual machine and configure the VM per your site's environment. Follow the instructions in the Slurm Quick Start Administrator Guide.
- Create an OS Image
  Create an image of the virtual machine configured per your site's requirements.
- Create an AWS Cloud Environment
  Create an AWS cloud environment in NavOps so that workload manager execution nodes can be deployed in the cloud. Provide information about the cloud environment such as region, OS image, subnet id, security group id and tags.
- Create an AWS Node Class
  Create an AWS Node Class so that workload manager execution nodes can be deployed in the cloud. A Node Class provides the infrastructure for the workload to run on. It is an instance type which defines benchmarked hardware. A Node Class exists in an Environment and consists of instance type, OS image, security group id, tags, SSH-keys and cloud-init script.
- Multi-Availability Zone Management on AWS
  Configure the cloud environment so that cloud nodes can be deployed in several Availability Zones.
Configure Microsoft Azure Cloud Scaling
Create the necessary components in Microsoft Azure that are required for cloud scaling and then add your Credential to NavOps before creating a cloud environment.
Windows Deployment on AWS and Azure
Requirements for deploying Windows cloud nodes.
Configure Google Cloud Platform Cloud Scaling
Create the necessary components in GCP that are required for cloud scaling and then add your Credential to NavOps before creating a cloud environment.
Configure Oracle Cloud Scaling
Create the necessary components in Oracle Cloud that are required for cloud scaling and then add your Credential to NavOps before creating a cloud environment.
Configure Openstack for Cloud Scaling
Create the necessary components in OpenStack that are required for cloud scaling and then add your credential to NavOps before creating a cloud environment..
Add, Register, and Deploy a Cluster
After creating a cluster, it has to be registered by installing the NavOps CLI and agent. You can add node configurations containing custom resources and scripts and deploy the cluster.
Node Configurations
Node configurations enable you to customize a node associated with a cluster. Compared to operations performed via the cloud-init script these configurations are managed on the WLM server increasing flexibility and security. Node configurations are supported for PBS, Grid Engine and Slurm clusters.
Deploy Cloud Nodes Manually
Deploy nodes manually (without the intervention of the workload manager) for testing purposes or when your site wants to keep nodes deployed for an indeterminate amount of time.
Undeploy Cloud Nodes Manually
Undeploy cloud nodes when they are no longer needed.
Automate Cloud Scaling for Accelerator
Create rules to automatically scale-up and scale-down deployment of nodes on the cloud.
Automate Cloud Scaling for PBS Professional
Create rules to automatically scale-up and scale-down deployment of nodes on the cloud. Use node configuration to customize a node associated with a PBS cluster.
Automate Cloud Scaling for Grid Engine
Create rules to automatically scale-up and scale-down deployment of nodes on the cloud. Use node configuration to customize a node associated with a Grid Engine cluster.
Automate Cloud Scaling for Slurm
Create rules to automatically scale-up and scale-down deployment of nodes on the cloud. Use node configuration to customize a node associated with a Slurm cluster.
Generate API Keys
NavOps uses API keys to connect the NavOps server, WLM agent, head node and cloud nodes. Each cluster has a dedicated API KEY, if you would like to generate additional key please use the following instructions. The key is generated and displayed once. Make sure you save it as you won't be able to access it again. The key is generated for a specific cluster. The key is set on the deployed nodes. It is also present in the NavOps agent on the head node and the MoMs. The default expiry date of the key is one year. When the key expires, you can generate another one and replace the key manually on the head node.
Cost Center
Cost centers provide visibility of expenditure to dynamically manage cloud operations and plan ahead. Cost centers are associated with node delpoyments. The limits defined in the cost center are checked before deploying nodes. The estimated costs accrued by cost centers are displayed visually as well as in a table.
Accounting
The accounting dashboard provides a set of reports containing charts and tables tracking Burn Rate, Core Count, Node Count and Total Cost for the Cloud connector, Cloud credentials, Cloud environment, Node class and Cluster datasets. These reports can be produced for a custom date and time range or a set of preset date and time ranges.
Known Issues

Multi-Availability Zone Management on AWS

Configure the cloud environment so that cloud nodes can be deployed in several Availability Zones.

Refer to the following AWS documentation if you are not familiar with AWS regions, Availability Zones, VPCs or subnets:

Deploying cloud nodes in multiple Availability Zones allows an HPC complex to distribute the load across a region and take advantage of AWS Spot Instances. Several prerequisites must be met before configuring the cloud environment:

An AWS virtual private cloud (VPC) must be created.
A subnet for each Availability Zone must be created and each subnet must belong to the same VPC.

Once these prerequisites are met, then it is as simple as providing a comma-separated list of subnets when the environment is created.

NavOps attempts to deploy cloud nodes in the first subnet in the list. If there is no availability in that subnet, then it will attempt to deploy cloud nodes in the next subnet in the list and will continue until a subnet is found where the cloud nodes can be deployed or until all subnets are exhausted (i.e. scaling fails). NavOps attempts to deploy all requested cloud nodes in a single subnet. Cloud nodes are not deployed across subnets. This same process is followed each scaling cycle until there is availability for the cloud nodes.

Example

10 cloud nodes are requested for scaling.

NavOps attempts to deploy all 10 nodes in subnet-014c5607b.
If there is no availability in subnet-014c5607b, NavOps attempts to deploy all 10 cloud nodes in subnet-0622f6467.
If there is no availability in subnet-0622f6467, NavOps attempts to deploy all 10 cloud nodes in subnet-05c352abff.
If there is no availability in subnet-05c352abff, then cloud scaling fails.