Fair multi-agent task allocation for large datasets analysis

Autor: Jean-Christophe Routier, Maxime Morge, Quentin Baert, Anne-Cécile Caron
Přispěvatelé: Systèmes Multi-Agents et Comportements (SMAC), Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 (CRIStAL), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS)-Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), Centrale Lille-Université de Lille-Centre National de la Recherche Scientifique (CNRS), This project is supported by the CNRS Challenge Mastodons.
Jazyk: angličtina
Rok vydání: 2018
Předmět:
Computer science
Distributed computing
Big data
02 engineering and technology
Negotiation
Task (project management)
[INFO.INFO-AI]Computer Science [cs]/Artificial Intelligence [cs.AI]
Artificial Intelligence
020204 information systems
0202 electrical engineering
electronic engineering
information engineering
MapReduce
Reducer
business.industry
Design pattern
Multi-agent system
Skew
Workload
Data structure
[INFO.INFO-MO]Computer Science [cs]/Modeling and Simulation
Human-Computer Interaction
Hardware and Architecture
[INFO.INFO-MA]Computer Science [cs]/Multiagent Systems [cs.MA]
020201 artificial intelligence & image processing
ACM: I.: Computing Methodologies/I.2: ARTIFICIAL INTELLIGENCE/I.2.11: Distributed Artificial Intelligence/I.2.11.3: Multiagent systems
business
Software
Information Systems
Zdroj: Knowledge and Information Systems (KAIS)
Knowledge and Information Systems (KAIS), 2018, 54 (3), pp.591-615. ⟨10.1007/s10115-017-1087-4⟩
Knowledge and Information Systems (KAIS), Springer, 2018, 54 (3), pp.591-615. ⟨10.1007/s10115-017-1087-4⟩
ISSN: 0219-1377
0219-3116
Popis: International audience; MapReduce is a design pattern for processing large datasets distributed on a cluster. Its performances are linked to the data structure and the runtime environment. Indeed, data skew can yield an unfair task allocation, but even when the initial allocation produced by the partition function is well balanced, an unfair allocation can occur during the reduce phase due to the heterogeneous performance of nodes. For these reasons, we propose an adaptive multi-agent system. In our approach, the reducer agents interact during the job and the task reallocation is based on negotiation in order to decrease the workload of the most loaded reducer and so the runtime. In this paper, we propose and evaluate two negotiation strategies. Finally, we experiment our multi-agent system with real-world datasets over heterogeneous runtime environment.
Databáze: OpenAIRE
Externí odkaz: