FWSS Network Capacity

Active datasets

—

not terminated

Daily FWSS gas

—

Daily target

70,560

Ggas/day · 5B × 4.9 × 2880

Headroom (today)

—

Ggas available for FWSS

Max datasets (today)

—

at current non-FWSS load

Daily gas usage vs target

Other

FWSS

Headroom

Non-FWSS (~15,327 Ggas today)

FWSS (—)

Headroom

Network scenario

How much FWSS headroom remains at different network loads?

Your datasets

number of active datasets

Avg pieces per dataset

drives provePossession gas via log₂(N)

FWSS gas/dataset/day

—

Mgas · computed from model

Non-FWSS gas

—

Ggas/day

Your FWSS gas

—

Ggas/day

Headroom left

—

Ggas/day remaining

Max datasets (budget)

—

at your avg piece count

▪ Non-FWSS▪ Your FWSS▪ Headroom

—

Dataset size distribution by piece count · mainnet

Size bucket	# datasets	Avg pieces	Gas/day	% of target
1 piece	225	1	283M	0.0004%
2–10 pieces	223	4	300M	0.0004%
11–100 pieces	155	61	333M	0.0005%
101–1k pieces	243	229	349M	0.0005%
1k–10k pieces	30	1,744	374M	0.0005%
>10k pieces	2	508,705	444M	0.0006%

Weighted average: ~323M gas/dataset/day · 475 active datasets = ~153 Ggas/day total

Weekly growth new datasets/week

Gas fee model: effective_gas_price = BaseFee + min(GasPremium, GasFeeCap − BaseFee) (Lotus source). Today: BaseFee ≈ 100 attoFIL, GasPremium ≈ 252k — SP tip dominates (99.9%). FIP-0115 (Draft): BaseFee would rise to ~90% of EGP. CDN datasets use 3 payment rails (+45M gas on createDataSet, +105M on terminateService). Data source: FOC Observer mainnet. addPieces model: gas = 228.72M + 78.20M × K (R²=0.9996, K=1..10 observed). K>310: single tx exceeds tipset budget → stuck forever. Overestimation burn: penalty for over-declaring GasLimit by >10%, scales with BaseFee.

⚠ Gas costs increased after FWSS v1.2.0 upgrade (23 Mar 2026):
createDataSet: ~250M (Nov 2025) → ~1,106M (Apr 2026) after the upgrade.
nextProvingPeriod: ~113M (Nov 2025) → ~153M (Apr 2026) as more datasets/CDN rails increase FilecoinPay settlement cost.
Always use live data from FOC Observer (click Refresh) for current values — hardcoded fallbacks reflect the post-upgrade era. Future contract upgrades may further change these values.