UbiOps version 2.15.0¶
On the 17th of March 2022 we have released new functionality and made improvements to our UbiOps SaaS product. An overview of the new functionality and changes:
On demand GPUs¶
We already provided GPU support but we expanded this functionality to be on-demand. That means that GPUs will now scale with your usage, within your subscription limits.
GPUs are only available on request. Please contact us if you would like to start using GPUs on UbiOps.
Please refer to the GPU deployments page for more information on using GPUs with UbiOps.
We added logging levels to our logs to make it easier to distinguish between different log types. UbiOps now supports two different log levels: info and error. Logs printed with, for example,
logging.info are marked with
level=info. While, logs printed with
logging.error and exceptions are marked with
level=error. The log levels are visible in the UbiOps WebApp.
Token and user specific metrics¶
The metrics system has been expanded to also show metrics specifically for a user, or for a service user (API token). You can find these metrics in the WebApp under Monitoring > Users & API tokens.
Soft subscription limit¶
When necessary, it is now possible to change your subscription limit into a soft limit. When enabled, you will be billed at the end of your billing cycle for the amount of compute you used outside of your normal subscription limits. This feature is only possible for approved paid customers. Please contact us if you need this feature as well.
We no longer install
gitby default in your deployment. If you need to use git within your deployment, you will have to install it manually using a ubiops.yaml.
We renamed the GPU instance type