Problem
Context: storage provider network. Clients delegate the storage of their file to a network of Storage Provides.
Goal
Design and implement a protocol that makes not answering retrieval requests irrational (ie, not profitable) for the Storage Providers.
Today Filecoin guarantees the persistent storage of the files that clients sent to the providers in the network. This is achieved via a market algorithm based on incentives (mining rewards), penalties and publicly verifiable proofs of storage. We want to design a protocol, eventually a market, that achieves the same guarantee for the retrieval of the stored files.
Data Availability | Data Retrievability | |
Storage | on-chain | off-chain |
Application | blockchain scaling (eg, full node/light clients paradigm, Rollups, ...) | guaranteed retrievability for data stored off-chain |
Security assumptions | a fraction of parties (providers and clients) are honest | (this work) providers and parties are rational |
Goal | data are available in the network (ie, a full node has the full block with all transactions) | guaranteed retrieval for the clients |
Outcome
We want to realize an MVP that solves the data retrievability problem. Steps:
Step 1: Design the protocol ✅
Step 2: Implementation ✅
Step 3: Filecoin integration and deployment (pending FVM) 🔁
Follow progress here: https://github.com/protocol/retriev
Team (2022)
Researchers: @Irene(DRI), @Luca Nizzardo, @Nicola
Engineers: external team (Sebastiano and Stefano from YOMI)
PM: external (0xJona)
Talks
- Presentation of our demo, by @Ireneduring the Retrieval Market Builders Demo Day - 17th Aug 2022
- “Filecoin Data Retrievability Consortium Update” : Talk by @Ireneat CryptoEconDay @FIL Austin - 9th June 2022
- “Filecoin Data Retrievability Consortium” : Talk by @Nicola at the Protocol Labs CryptoEconDay - 22nd April 2022
Links
💻 Code repo: