FWS & PDP Exploration

Questions

Q: How to communicate with Curio to manage PDP ProofSets, Roots and file data? Is there any API we can use? Our backend will run on a different machine than the Curio SP.

Discussion with FilOz:
https://space-meridian.slack.com/archives/C0838QS1CVC/p1747245913222809

pdptool was a tooling thing for making the things connected (edited)
not a final lib
it will be wrapped behind apis

Raw Notes & Links

  • PDP with Payments using a custom made PDP Service:https://github.com/FilOzone/pdp/pull/146/files
    • Limitations:
      • proofSetDeleted not implemented (Payment rail termination; not needed in MVP)
      • rootsScheduledRemove not implemented
      • Hard-coded fees
        PROOFSET_CREATION_FEE = (1 * 10 ** tokenDecimals) / 10; 
        // 0.1 USDFC
        
        RATE_PER_MIB_PER_EPOCH = (1 * 10 ** tokenDecimals) / 10;
        // 0.1 USDFC per MiB per epoch

        That rate means 8,640 USDFC per MiB per month 😱

Resources

PDP Explorer

User Documentation

How to download from PDP

https://space-meridian.slack.com/archives/C08TC8NMM17/p1747737521881809
curl https://yablu.net/piece/baga6ea4seaqkzso6gijktpl22dxarxq25iynurceicxpst35yjrcp72uq3ziwpi -o data.pdf

Development Synergies

There are several development roadmaps that we as a team can allocate resources to. They are structured roughly into three different categories:

  • 0K Spark
  • FWS for retrievability
  • FilStore

The development of the FWS service for Retrievability will involve getting familiar with the FWS services for payments and the Service Level Indicator, and how to build a Service Level Agreement. The work on FilStore will include writing a Service Level Agreement. This means that development work on FWS for retrievability will potentially speed up the implementation of the SLA for Filstore. Know how that is acquired through implementing the FWS will be transferable to the work that is required to implement SLA and FWS communication for FilStore.