An overview of system health and performance
Data driven governance requires data. Deriving insights from data requires maps and models. This site contains a series of notebooks that pull live data from the Filecoin network, run it through models, and then visualize key metrics across a set of dashboards. This can provide insights to guide FIP discussions and stakeholder decisions to improve the Filecoin network.
As we develop more models, metrics, and dashboards we will add them to this page. This will provide a growing body of knowledge and key metrics to inform the Filecoin Improvement Proposals.
In this notebook we perform Exploratory Data Analysis (EDA) on FIL's gas fee mechanism.
The
goal is to observe the gas fee as a signal and attempt to understand what may be driving
it.
We perform Fourier
Transformations on the minute aggregated derived_gas_outputs data.
Builds on Gas EDA V1 and performs Fourier Transformations on
message_gas_economy
data. Additionally, chain_rewards, chain_powers, and
chain_economics event level data is analyzed and Granger Causality is
performed
on the signals.
Builds on the previous two notebooks second aggregated derived_gas_outputs
data
was pulled for July 2021 as well as the actor types and methods.We created a mapping
of actor_types and methods (as methods are not unique across actor types) downloaded all
July data per actor_type and created histograms of gas_used
over
method type. We then combined the derived_gas_outputs,
chain_rewards, chain_powers, and chain_economics
tables and performed VAR
and Granger Causality
Analysis.
The goal of this notebook is to explore the impact of the sector expiration wave in April/May 2021, especially wrt V1 proofs.
With the Baseline Target Threshold passed Filecoin block rewards are now primarily distributed according to the Baseline Minting policy. This provides ongoing incentives to maintain a baseline of storage capacity on the Filecoin network. There are two articles exploring what crossing the Baseline Target Threshold means and how Filecoin block rewards work:
This notebook continues to track network power wrt the baseline minting threshold for ongoing monitoring.
In this notebook, we will evaluate the daily performance of our gas usage method percentage systems identification VAR(X) model.
This notebook examines if Filecoin's base fee trends towards stationarity, as hypothesized by Dr. Barnabé Monnot's work on EIP 1559.
This notebook examines if a structural break occured following China's new cryptcurrency regulations.
The Sector State Metrics Dashboard shows missing Proofs of Space Time (PoST) over time. This is important because missing PoST implies unreliable storage services. By monitoring the % of PoST faults we can verify that the protocol is functioning as intended and terminating faulty sectors and that the % of sector faults is staying within an acceptable range.
This dashboard can show us the lifetime of sectors. This is important to anticipate when sectors will expire, to compare the ratio of sector expirations vs activations and to understand the average lifetime of sectors. When sectors expire miners can either reinvest the FIL and equipment into a new sector, or sell them on the market. If there are more sector activations than expirations that puts positive price pressure on FIL and mining equipment. If there are more expirations than activations that might show that miners find storage provision to be unprofitable and are waiting for market conditions to change or dumping capital (FIL and equipment) on the market.