Abstract: This paper proposed a decentralized mechanism to solve resource allocation for computational grid where sellers have finite production capacities and a cost structure composed of a fixed overhead cost and a constant marginal cost. The RB2-MCDA mechanism is that agents use risk-based2 bidding strategy for transactions in MCDA. Such mechanism is required when a system consists of multiple self-interested stakeholders of which each possesses private information that is relevant to solving a system-wide problem. The experimental results show that the RB2-MCDA mechanism achieves high efficiency.

Key words: computational grid, risk-based2 bidding strategy, multi-unit continuous double auction

摘要： 针对计算网格资源的特点，提出一种基于风险策略的多单元连续双向拍卖的网格资源分配机制，实现对网格资源灵活有效的管理。首先，介绍了基于多单元连续双拍卖的网格资源分配框架。其次，针对计算网格资源的有限性，提出了RB2-MCDA机制。RB2-MCDA机制是在多单元连续双向拍卖中，代理采用Risk-Based2策略进行资源交易。Risk-Based2策略是一种基于风险行为的代理策略。实验结果表明，在不同规模的有限资源的计算网格中采用RB2-MCDA机制能够实现较高的资源分配效率，当资源需求量接近供给量时，分配效率超过99%。

关键词: 计算网格, Risk-Based2策略, 多单元连续双向拍卖