Carolina Cloud Storage

Carolina Cloud provides managed S3-compatible object storage through Wasabi. This gives you a persistent storage layer that survives instance deletion, making it ideal for keeping datasets, results, and reference files between ephemeral compute sessions.

Provisioning a bucket

From the Storage section of the dashboard, click Provision Storage. Carolina Cloud creates:

• A dedicated Wasabi S3 bucket
• An IAM user with scoped permissions (read/write access to your bucket only)
• Access key and secret key credentials

These are displayed in the dashboard and can be auto-filled into Marimo and RStudio Server instances at creation time.

Using your bucket

From the dashboard

The built-in file browser lets you view folders and files in your bucket without leaving the console. Navigate to Storage and browse your bucket contents.

From an instance

When creating a Marimo or RStudio Server instance, click Use my Carolina Cloud bucket in the S3 Integration section. Your credentials and endpoint will be auto-filled.

Your bucket is automatically mounted at ~/ccloud-s3/ using s3fs. This means you can work with your S3 storage exactly like a normal folder — no aws commands required:

# Browse your bucket
ls ~/ccloud-s3/

# Copy a file into your bucket
cp results.csv ~/ccloud-s3/experiments/

# Read data directly
head ~/ccloud-s3/datasets/sample.csv

# Use it from R, Python, etc. — just reference the path
# e.g. pd.read_csv("~/ccloud-s3/datasets/sample.csv")

Under the hood, ~/ccloud-s3/ is backed by s3fs-fuse, which translates file operations into S3 API calls transparently. Any file you write there is persisted in your Wasabi bucket immediately.

The aws CLI is also pre-configured if you prefer it:

aws s3 ls s3://your-bucket-name/
aws s3 cp results.csv s3://your-bucket-name/experiments/

From your local machine

Use any S3-compatible client with the credentials from the dashboard:

export AWS_ACCESS_KEY_ID="your-access-key"
export AWS_SECRET_ACCESS_KEY="your-secret-key"
export AWS_ENDPOINT_URL="https://s3.us-central-1.wasabisys.com"

aws s3 ls s3://your-bucket-name/

Pre-filling instances

Note

With the automatic ~/ccloud-s3/ mount, pre-filling is often unnecessary — your bucket data is already accessible as a filesystem. Pre-filling is still useful when you need data copied to local disk for performance-sensitive workloads.

When creating an instance, specify aws_prefill_bucket to download data from your bucket into ~/s3-prefill/ at startup:

curl -X POST https://console.carolinacloud.io/api/instance/ \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "resource_type": "container",
    "flavor": "rstudioserver",
    "n_vcpus": 8,
    "mem_gib": 32,
    "disk_size_gib": 500,
    "aws_access_key_id": "your-access-key",
    "aws_secret_access_key": "your-secret-key",
    "aws_endpoint_url": "https://s3.wasabisys.com",
    "aws_prefill_bucket": "your-bucket-name/datasets/"
  }'

The download happens during provisioning. When your instance is ready, the data is already on disk.

Deleting your bucket

From the Storage section, click Delete Storage. This permanently deletes:

• All objects in the bucket
• The bucket itself
• The associated IAM user and access keys

This action cannot be undone.

Pricing

Warm S3 storage is billed at $9/TB/month, prorated hourly, with no egress fees. See Pricing & Billing for the full breakdown and the Storage Deletion Policy for how Wasabi’s 30-day minimum retention is handled.