Transaction Fees Optimization in the Ethereum Blockchain

Introduction

Ethereum's blockchain relies on transaction fees ("gas") to prioritize and process user transactions. This paper explores methods to determine the minimum gas price a user should pay to ensure their transaction is mined within a specified time frame with a given probability.

Key Concepts

• Gas Price (gp): Fee per unit of gas (denominated in GWei).
• Gas Limit (gl): Maximum gas a transaction can consume.
• Block Gas Limit (BL): Maximum gas allowed per block (~12.5M gas as of 2021).

Ethereum Blockchain Mechanics

1. Transaction Processing

• Miners select transactions based on gas price, prioritizing higher fees.

• Transactions must meet validity criteria:

  • Sufficient sender balance.
  • Correct nonce sequence.
• Pending Transactions: Wait in a mempool until mined.

2. Gas Mechanism

• Smart Contracts: Execute code requiring gas. Fees = gasPrice × gasConsumed.
• Miners set a minimum gas price threshold (LGP). Transactions below this threshold are ignored.

Challenges in Gas Price Optimization

Users face a trade-off:

• Overpaying: Ensures quick processing but wastes Ether.
• Underpaying: Risks delays or stuck transactions.

Existing Solutions

• ETH Gas Station: Predicts gas prices statistically but lacks configurability for time/probability.
• Machine Learning Models: Predict lowest gas prices but ignore pending transactions’ state.

Proposed Method: Monte-Carlo Simulation

Step 1: Scenario Generation

Generate SC scenarios simulating:

• Future transaction broadcasts (W').
• Block mining times (V').

Step 2: Probability Computation

For each scenario sc, compute:

• Mining Probability (P(V_tr(T))): Likelihood transaction tr is mined within time T.
• Binary Search: Find the minimal gp_tr satisfying P(V_tr(T)) ≥ α.

Algorithm Overview

1. Input: Target time T, probability α, historical data (V, U, W).
2. Output: Optimal gp_tr.

Experimental Results

Data Collection

• Period: Nov 2019 (5M transactions).
• Cleanup: Removed invalid/unbroadcast transactions.

Performance Metrics

| Metric | Value (Delay=0ms) | Value (Delay=2500ms) |
|----------------------|-------------------|-----------------------|
| Avg. Precision (α=0.7) | 76% | 71% |
| Computation Time | ≤2s per prediction | |

Key Findings

• The method reliably predicts gas prices within ±2% of desired probability.
• Introduces a configurable delay parameter (2500ms) to align predictions with real-world mining delays.

FAQs

1. How does gas price affect transaction speed?

Higher gas prices increase mining priority. A transaction with gp in the top 10% of the mempool is typically mined within 1–2 blocks.

2. Can I cancel a pending transaction?

Yes, by sending a replacement transaction with the same nonce and higher gp. Success depends on initial gp and mempool congestion.

3. Why do some transactions get stuck?

• Insufficient sender balance.
• Preceding transaction (prec_μ) not yet mined.

👉 Learn more about Ethereum gas dynamics

Conclusion

The proposed Monte-Carlo method provides configurable, reliable gas price predictions by:

1. Simulating future transaction flows.
2. Accounting for miners’ gas price thresholds.
3. Optimizing for user-defined time/probability constraints.

Future Work: Integrate real-time mempool data to refine predictions during network congestion.

👉 Explore advanced blockchain analytics