paper-with-me

Papers

Parallel Execution Fee Mechanisms

2024-10-12 · Abdoulaye Ndiaye

This paper investigates how pricing schemes can achieve efficient allocations in blockchain systems featuring multiple transaction queues under a global capacity constraint. I model a capacity-constrained blockchain where users submit transactions to different queues -- each representing a submarket with unique demand characteristics -- and decide to participate based on posted prices and expected delays. I find that revenue maximization tends to allocate capacity to the highest-paying queue, whereas welfare maximization generally serves all queues. Optimal relative pricing of different queues depends on factors such as market size, demand elasticity, and the balance between local and global congestion. My results have implications for the implementation of local congestion pricing for evolving blockchain architectures, including parallel transaction execution, directed acyclic graph (DAG)-based systems, and multiple concurrent proposers.

📄 PDF Abstract BibTeX arXiv:2410.09555

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

CONCERTO: Complex Query Execution Mechanism-Aware Learned Cost Estimation

2024-12-01 · Kaixin Zhang, Hongzhi Wang, Kunkai Gu, ZiQi Li 외

With the growing demand for massive data analysis, many DBMSs have adopted complex underlying query execution mechanisms, including vectorized operators, parallel execution, and dynamic pipeline modifications. However, t…

Graph AttentionPrediction

Automatic Task Parallelization of Dataflow Graphs in ML/DL models

2023-08-22 · Srinjoy Das, Lawrence Rauchwerger

Several methods exist today to accelerate Machine Learning(ML) or Deep-Learning(DL) model performance for training and inference. However, modern techniques that rely on various graph and operator parallelism methodologi…

Beyond Speedups: Hardware-Aware Evaluation of Evolutionary Algorithms on GPUs

2026-01-26 · Xinmeng Yu, Tao Jiang, Ran Cheng, Yaochu Jin 외 arxiv

Evolutionary algorithms (EAs) are increasingly executed on graphics processing units (GPUs) to exploit population-level parallelism. This shift changes the resource model under which EAs are designed and evaluated. Howev…

DynTaskMAS: A Dynamic Task Graph-driven Framework for Asynchronous and Parallel LLM-based Multi-Agent Systems

2025-03-10 · Junwei Yu, Yepeng Ding, Hiroyuki Sato

The emergence of Large Language Models (LLMs) in Multi-Agent Systems (MAS) has opened new possibilities for artificial intelligence, yet current implementations face significant challenges in resource management, task co…

ManagementScheduling

FlashDiff: Efficient Regional Execution and Scheduling for Diffusion Model Serving

2026-07-13 · Yaqi Qiao, Ping He, Songrun Xie, Ayush Barik 외 arxiv

Diffusion models have become the central backbone for modern image, video, and audio generation, but their efficient service remains a challenge. Unlike autoregressive decoding, diffusion inference repeatedly updates hig…

Audio Generation