Logo

    CryptoNet is a Protocol Labs initiative.

    Scaling FIL+ VC Allocation

    Status
    šŸŸ£
    DRI
    Team
    šŸ”® Protocol Opportunities
    Date
    May 1, 2023 ā†’ June 30, 2023
    Goal
    Quarter
    2023Q3
    Subgoal
    Implement a first version for concrete benchmarks

    šŸšØProblem

    The FIL+ allocation strategy is slow, costly and does not scale onchain.

    šŸ’”Goal

    Clients can create a single allocation for data that will span multiple sectors.

    The primary goal is to reduce theĀ workload/cost on the clientĀ and for the SP as well by claiming multiple allocationsĀ at once.

    šŸš€Ā Mechanism

    We propose the clientĀ use a commitmentĀ to represent a set of Piece CIDs (CommPs) that the client wishes to put on Filecoin.

    The client can then do aĀ single allocation for this set of CommPs.

    The SP can claim set of commPs eitherĀ incrementallyĀ (one by one) or by batches.

    Expected gas cost improvements:

    image

    šŸ“šĀ Documents

    • FIP discussion https://github.com/filecoin-project/FIPs/discussions/708
      • Internal draft: https://www.notion.so/pl-strflt/FIP-Discussion-d45d24322ad14a22a4484e79da70c757
    ā€£
    Gas improvements
    ā€£
    Overall cost for syscall
    • 10gas/ns is the translation factor
    • 20k overhead per syscall
    • 0.6gas per byte
    ā€£
    Overall cost for storage
    ā€£
    1300 gas / byte (non indexed storage), 3400 / byte indexed
    ā€£
    KZG cost (syscall): approx 10M
    ā€£
    pairing check 1ms computation ā†’ 10M gas
    ā€£
    373microseconds ā†’ 373k gas
    ā€£
    KZG takes 48 bytes proofs + indices + values
    • X * 32 * 2 * 0.6 = nothing, 2k for 64 opening
    ā€£
    KZG Cost storage: 48B
    ā€£
    Merkle Tree cost for 1 opening:
    ā€£
    320B * 1300 = 416k
    ā€£
    Merkle Tree cost for 64 opening:
    ā€£
    320*64*1300 = 26M