Automatic Resizing

With Spell, you don't have to worry about your runs running out of disk. If you are on our Teams or Enterprise plans and using an external cluster, Spell will automatically resize your disk when it is nearing capacity without interrupting your run.

How it works - AWS External

AWS disks can resize slowly and only once every six hours. These are limitations of AWS, not Spell. To account for this, the threshold for AWS disk resizing can be lower than the threshold for GCP to account for a resize operation taking longer, and the resize factor (how much larger the disk gets) is much higher. Since resizes are limited to once every 6 hours, resizing multiplies the disk size by a factor of 10. Larger disk size increases cost, but Spell recycles machines that are no longer in use to ensure you are using machines efficiently and keeping costs down. Effectively, this means that runs will always start with the the disk size you specified during machine type creation and resize to larger disks as needed.

How it works - GCP External

GCP disks can resize quickly and multiple times in succession. Using any more than 80% of your disk on an external GCP cluster will trigger a resize doubling that disk's capacity. If you need more disk, the disk will keep resizing. There are no limits on the number of times you can resize the disk on a GCP cluster. Note that GCP will charge more for the increased disk, but Spell recycles machines that are no longer in use to ensure you are using machines efficiently and keeping costs down. Effectively, this means that runs will always start with the disk size you specified during machine type creation and resize to larger disks as needed.

Note

In order to use this feature on GCP, your cluster must be up to date. To update your cluster, please run spell cluster update