Configure an Environment Pool with Autoscaling

You can configure an environment pool with autoscaling using Oracle Cloud Infrastructure instance pools.

An instance pool defines a set of compute instances that's managed as a group. Managing instances as a group enables you to efficiently provision instances and manage the state of instances.

The base of the scaling decision is the environment pool utilization. The utilization of an environment pool is calculated as the average utilization of all environments within that pool. The utilization of an individual environment is calculated based on the number of robot instances currently running in that environment and the number ofinstances queued for the same environment. For example, if an environment is currently processing 1 bot instance and 1 more execution is waiting in the queue, then the utilization of that environment becomes 200%. Hence, if Environment1 has utilization of 200% and Environment 2 has utilization of 300%, then the environment pool utilization will be the average of both, which is 250%.

You must complete a one-time setup to configure Oracle Cloud Infrastructure instance pools to define a reusable infrastructure blueprint that can instantiate instance(s) on demand. The RPA platform will then use this to dynamically scale robot agent environments based on real-time workload demands. RPA monitors the robot workload and based on workload demand it will scale-out and scale-in the robot agent environments.

This solution supports:

  • Auto-provisioning of VMs using Oracle Cloud Infrastructure managed instance pools.

  • Dynamic elasticity, where VMs are added or removed as needed. The robot agent environment will automatically scale-out or scale-in based on the utilization of the environment pool.

  • Automatic agent registration, allowing new environments to auto-join RPA environment pools.

The following topics describe the required configuration details for: