Google Drive API via Service Account - Storage quota exceeded

I have a google service account (Owned by me), which has been given access to another users Google Drive Account (Owned by someone else - Standard 15GB storage). They granted full access to a single folder under their G-Drive account, and I can see that the SP has full access to the folder when using the googleapiclient python package.

I have a process that lists all the files within that folder, and overwrites some of them based on new/incoming data from an external source. Here is what I see:

• Files that are very small (few KB to few MB) let me overwrite them over and over again, with no issues. I do not see any version history, which I kind of expected to see.

• Files over 100MB seem to have an issue with hitting that quota. It doesn't matter if the file is already sitting on the server and I am overwriting it, or if I just remove the file completely and try and upload it. Oddly enough, I don't hit this every time and I have no issue actually uploading the file from Python as I am using multipart/resumable.

• When I get this quota error, I can still log into the google UI and manually upload files to that drive just fine. No errors.

• When I log into the google drive account from a UI, I can see there are 0 bytes used for "Storage", even if I have 15 or so small-ish files in "My Drive".

So my questions are:

1. Do Service Accounts have a transfer quota? I know for sure the Drive is not full, but I still get that quota limit with bigger files.

2. Are there differences between "My Drive" and "Storage" that I need to account for, such as impersonating a given user?

Thanks!

This is what I have done, which seems to be working, so far:

def get_service_user(api_name, api_version, scopes):
creds = None

if os.path.exists("token.pickle"):
    with open("token.pickle", "rb") as token:
        creds = pickle.load(token)

if not creds or not creds.valid:
    # Assuming here that once we have done the "grant" once as a real user, this file can then be moved to where its running, and update automatically.
    # User will need to run script locally and collect token.pickle file / give to person automating this thing.
    # Or if this runs in a UI, the back end process can just store for a given user.
    if creds and creds.expired and creds.refresh_token:
        creds.refresh(Request())
    else:
        flow = InstalledAppFlow.from_client_secrets_file("credentials.json", scopes)
        # Can't run local server on Databricks. If the user changes, you must run this locally, authenticate with google, and store this file back in DBFS (location above)
        # Uncomment this blow "creds" section to run locally.
        # creds = flow.run_local_server(port=0)
        print("The token can't be refreshed.  Please run this code locally and generate a new pickle file")
        exit(0)

    # Save the credentials for the next run
    with open("token.pickle", 'wb') as token:
        pickle.dump(creds, token)

# return Google Drive API service
service = build(api_name, api_version, credentials=creds)
return service