Runtime environment usage

Runtime usage is calculated by the number of capacity unit hours (CUH) consumed by an active environment runtime. Only the free environment runtime doesn't track the amount of CUH that it consumes. All other environment runtimes do.

CUH calculation for default environments

The rate of capacity units per hour consumed by a default Anaconda environment is determined by the environment definition of the runtime and the number of users in a project using one or more runtimes.

This table shows the capacity units per hour calculation for default Anaconda environments.

Environment type Capacity units per hour
1 vCPU and 4 GB RAM 0.50
2 vCPU and 8 GB RAM 1
4 vCPU and 16 GB RAM 2
8 vCPU and 32 GB RAM 4
16 vCPU and 64 GB RAM 8

For example, a notebook runtime created by using the Default Anaconda XS environment definition with 2 vCPUs will consume 1 CUH if it runs for one hour. If you have a project with 7 users working on notebooks 8 hours a day, 5 days a week, all using the Default Anaconda XS environment, and everyone shuts down their runtimes when they leave in the evening, runtime consumption is 5 x 7 x 8 = 280 CUH per week.

The CUH calculation is more complex when different environment definitions are used to run notebooks in the same project, for example, Python 3.5 and Python 2.7 environments, some customized ones, and if users have multiple active runtimes, all consuming their own CUHs. Additionally, there might be notebooks, which are scheduled to run during off-hours, and long-running jobs, likewise consuming CUHs.

CUH calculation for Spark environments

The rate of capacity units per hour for a Spark environment is calculated based on the hardware configuration size of the driver, and the number of executors and their size.

driver size + (number of executors * executor size)

This table shows the capacity units per hour calculation for Spark environments.

Environment type Capacity units per hour
Driver with 1 vCPU and 4 GB RAM
1 - 10 executors with 1 vCPU and 4 GB RAM
1 - 10 executors with 2 vCPU and 8 GB RAM
Driver: 0.50
Each executor: 0.50
Each executor: 1
Driver with 2 vCPUs and 8 GB RAM
1 - 10 executors with 1 vCPU and 4 GB RAM
1 - 10 executors with 2 vCPU and 8 GB RAM
Driver: 1
Each executor: 0.50
Each executor: 1

For example, a notebook runtime created for a Spark environment with a driver size of 2 vCPUs (which consumes 1 capacity unit per hour) and 4 executors each 1 vCPU (which consumes 0.5 of a capacity unit per hour) in size, will consume 1 + (4 * 0.5) = 3 capacity unit hours.

CUH calculation for GPU environments

The rate of capacity units per hour consumed by a GPU environment is determined by the hardware configuration.

This table shows the capacity units per hour calculation for GPU environments during beta.

Environment type GPUs Capacity units per hour
1/2 x NVIDIA TESLA K80 1 1
1 x NVIDIA TESLA V100 1 4

During the beta phase, you can only consume a maximum of 100 CUHs a month. After this amount has been consumed, you must wait for the next month before you can consume capacity units for a GPU environment again. During beta, you will not be charged for the 100 capacity units you consumed in the month.

Runtime capacity limit

You are notified when you're about to reach the runtime capacity limit for your Watson Studio service plan. When this happens, you can:

  • Continue using the small default environment runtime (with 1 vCPU and 4 GB RAM) you have for free.

    You can create any number of these small environment runtimes but only one free environment can be active at any one time.

  • Stop active runtimes you don't need.
  • Upgrade your service plan. For up-to-date information, see the Watson Studio pricing plans.

The CUH counter continues to increase while a runtime is active so stop all runtimes you aren't using. If you don't explicitly stop a runtime, the default runtimes are stopped after an idle timeout of 1 hour for Lite plan users and 3 hours for paid plan users. Spark environment runtimes are stopped after an idle time of 3 hours. During this idle time, you will continue to consume CUHs for which you are billed.

Usage limitation for the free environment

Only one free environment can be active per account at any one time. When you start a notebook with the free environment runtime, you are notified if the free runtime is already active for your account. This can happen, for example, when you create and customize several of these small runtime environments in one project or even across projects you created, or when multiple users in the same project share environment definitions.

A free runtime doesn't run infinitely. It is automatically shut down after 12 hours, regardless of activity.

If the free runtime is already active:

  • As account admin of the project, you can see which runtimes are active for your account across all of your projects from the Environment Runtimes page and stop the other free runtime. Then try opening your notebook again. You can also see which runtimes are active and stop runtimes from the Environments page of your projects.
  • As project admin of the project, you can stop the other active free runtime from the project's Environments page.
  • You can change the environment of your notebook to an environment which consumes capacity units from the Assets page of your project:
    1. Select your notebook from the Notebooks section.
    2. Click Actions > Change environment and select an environment which consumes capacity units.

Track runtime usage for a project

You can view the environment runtimes that are currently active in a project, and monitor usage for the project from the project's Environment page.

Runtimes started for scheduled notebook jobs are marked as notebook jobs in the Tool column in the active environment runtimes list.

Track runtime usage for an account

The CUH consumed by the active runtimes in a project are billed to the account that the project creator has selected in his or her profile settings at the time the project is created. This account can be the account of the project creator, or another account that the project creator has access to. If other users are added to the project and use runtimes, their usage is also billed against the account that the project creator chose at the time of project creation.

You can track the runtime usage for an account on the Environment Runtimes page if you are the IBM Cloud account owner or administrator.

To view the total runtime usage across all of the projects and see how much of your plan you have currently used, choose Manage > Environment Runtimes.

A list of the active runtimes billed to your account is displayed. You can see who created the runtimes, when, and for which projects, as well as the capacity units that were consumed by the active runtimes at the time you view the list.

Learn more