Storage of an AzureML compute instance

[prisma idol]

The image shows the file system of a compute instance.

1. This link says I have 120GB of OS Disk ( https://learn.microsoft.com/en-us/azure/machine-learning/concept-compute-instance?view=azureml-api-2#compute-target ), but in reality I only have ~60GB free. Is there anything I can do to have more?

2. From the screenshot above, can someone please help me explain what these directories are (HDD or SSD?, data lost after each start/stop?) , and their reccommended usage?
   /dev
   /dev/shm
   /run
   /run/lock
   sys/fs/cgroup
   /mnt

3. I have read that I could store training data (images, about 50GB) at /home/azureuser. However, I will barely have enough space. I don't want to download the data from blobstorage to compute instance everytime. What should I do?

Thank you very much

[sterile idol]

Most mounts are irrelevant. The /mnt should be a temporary disk
https://learn.microsoft.com/en-us/azure/virtual-machines/managed-disks-overview#temporary-disk

[sterile idol]

you can use the fileshare in your AML workspace as a shared storage https://learn.microsoft.com/en-us/azure/machine-learning/concept-compute-instance?view=azureml-api-2#accessing-files