🛠 Storage & Retrieval on Filecoin Part 1
summary
Two concrete flows: a direct deal path where a contract emits a proposal and storage providers running Boost fetch a CAR file, publish the deal on-chain and seal sectors (sealing takes hours; mainnet sectors are 32–64 GB, localnet demo uses 2 KB), and an aggregator path via Filecoin Data Tools using Edge and Delta that bundles many files, produces an inclusion proof for each item, creates an aggregate CAR, and then Delta makes the storage deal. Filecoin hosts ~12–13 exabytes with ~3,500 providers; retrieval tools shown include Lassie and the deal client contract examples.
AI-written summary · about
chapters (6)
- 0:00Introduction to FVM storage
- 2:34Storage approaches explained
- 7:58Technical storage architecture
- 12:25Direct deal demo
- 20:19Aggregator storage demo
- 29:39Q&A and concluding remarks
related talks
29mFVM Intro: Build an AI Dataset Marketplace
28mOpening Remarks / Filecoin: What's Now Possible
15mThe State of Filecoin
14mFilecoin - Building with the Filecoin Onchain Cloud
21mFilecoin Workshop
12mWhy Do We Need a Decentralized Storage System?
26mCatching Up with the New Data Economy
32mWeb3 Storage and NFT Storage