Add to Chrome
Add to Firefox
Add to EdgeEfficient Resource Scheduling for Distributed Infrastructures Using Negotiation Capabilities
Feb 13, 2024
In the past few decades, the rapid development of information and internet technologies has spawned massive amounts of data and information. The information explosion drives many enterprises or individuals to seek to rent cloud computing infrastructure to put their applications in the cloud. However, the agreements reached between cloud computing providers and clients are often not efficient. Many factors affect the efficiency, such as the idleness of the providers' cloud computing infrastructure, and the additional cost to the clients. One possible solution is to introduce a comprehensive, bargaining game (a type of negotiation), and schedule resources according to the negotiation results. We propose an agent-based auto-negotiation system for resource scheduling based on fuzzy logic. The proposed method can complete a one-to-one auto-negotiation process and generate optimal offers for the provider and client. We compare the impact of different member functions, fuzzy rule sets, and negotiation scenario cases on the offers to optimize the system. It can be concluded that our proposed method can utilize resources more efficiently and is interpretable, highly flexible, and customizable. We successfully train machine learning models to replace the fuzzy negotiation system to improve processing speed. The article also highlights possible future improvements to the proposed system and machine learning models. All the codes and data are available in the open-source repository.
