Appendix D: Erasure Coding Explained
A simplified explanation of how Anvil uses Reed-Solomon erasure coding to provide high durability with low storage overhead.
A simplified explanation of how Anvil uses Reed-Solomon erasure coding to provide high durability with low storage overhead.
An overview of the Worka platform as a secure, local-first host application that runs sandboxed AI Packs.
A breakdown of the three pillars of a pack - the UI (React Views), the backend logic (MCP Servers & Tools), and the manifest (aip.json).
A detailed look at how Anvil nodes discover each other and maintain cluster state using a SWIM-like gossip protocol over QUIC.
An exploration of Anvil's vision for unifying storage and compute, allowing peers to execute jobs directly on the data fabric.
A detailed look at how Anvil achieves durability and efficiency using erasure coding, content addressing, and shard placement.
A detailed look at Anvil's metadata architecture, including the global vs. regional database split and the use of advanced PostgreSQL features.
A high-level overview of the Anvil architecture, its guiding principles, and the core technologies it is built on.
A step-by-step walkthrough of how Anvil handles a PutObject (write) and GetObject (read) request.