An Adaptive Compute-and-Wait Proof-of-Work Framework for Energy-Efficient and Scalable Blockchain Consensus
Year: 2026
Type: article
Abstract: Blockchain consensus algorithms must balance security, scalability, fairness, and energy efficiency. Compute-and-Wait Proof-of-Work (CW-PoW) organizes mining in short compute rounds. A round closes when NbrS solutions are visible, and only their authors continue. This paper analyzes five downsides of CW-PoW: latency inflation, parameter sensitivity, idle energy, communication overhead, and tie-breaking. Five remedies are specified and costed: adaptive parameter tuning, pipelined rounds with a commit-reveal round identifier, useful-work substitution, duty-cycling, and committee finalization. The finalization layer is positioned against existing constructions, and the security arguments are proof sketches. An event-driven model, checked against a threaded prototype, is used for the evaluation. At the same mean block time as PoW, CW-PoW consumes 71% to 95% less energy per miner and block for 10 to 1000 miners. Its 99th percentile of the block time is 61% to 70% shorter, and its mean block time does not grow with the number of miners. The fairness ratio of every hash-power decile lies between 0.85 and 1.17 for 100 and 1000 miners if NbrS≤Nm/4. The overlap of the header precomputation with hashing saves 5.7% of the energy, and the committee adds 1.0% to the latency.
Source: Network
Author: Mostefa Kara
Institution: University of Eloued
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,207
OpenAlex ID: W7235929544
