Storage#

#Your data lives on the blockchain, not on someone's server

Unlike traditional cloud storage, Rabbithole doesn't store your files on company-owned servers. Your encrypted data is stored directly in your own canister on the Internet Computer — a decentralized blockchain network.

#Each user gets their own canister

When you start using Rabbithole, a personal canister is deployed just for you. This canister belongs to you — after deployment, Rabbithole removes itself from the controller list, making you the sole owner.

Learn more about this in Data Sovereignty.

#What are canisters?

Canisters are smart contracts on the Internet Computer. Think of them as programs that run on a decentralized network of independent computers. They:

• Execute code transparently (anyone can verify)
• Store data in tamper-proof memory
• Cannot be shut down by any single entity
• Run 24/7 without downtime

Your personal canister handles both:

#What happens to my data if Rabbithole disappears?

Your data persists in your canister as long as it has cycles (the ICP computation currency). Since the code is open-source, you can:

1. Deploy your own frontend
2. Point it at your existing canister
3. Access your files

---

#Technical Details

Click to expand technical details

#Stable Memory & Memory Regions

File fragments are stored in ICP Stable Memory — persistent storage that survives canister upgrades. Each fragment is managed through Memory Regions, which provide:

• Efficient blob allocation and deallocation
• Random access by fragment index
• Contiguous storage for streaming reads

#Storage layout

#Capacity

• Each canister: up to 400 GB of stable memory
• Fragment size: ~1.9 MB plaintext (~1.9 MB + 28 bytes encrypted)
• A single canister can store ~200,000 fragments

#Data integrity

• SHA-256 checksums verify fragment integrity on download
• Fragments are materialized before memory operations to prevent use-after-free