Mutualistic Compute Continuum: A Network Economics Analysis
Mutualistic Compute Continuum: A Network Economics Analysis
复制标题
DOI：
10.1109/iccworkshops57953.2023.10283544
复制
发表时间：
2023-05
期刊：
影响因子：
--
通讯作者：
Aisha B. Rahman;Md. Sadman Siraj;E. Tsiropoulou;S. Papavassiliou
中科院分区：
文献类型：
--
作者：
Aisha B. Rahman;Md. Sadman Siraj;E. Tsiropoulou;S. Papavassiliou
在数据驱动的变革时代，由边缘、雾和云组成的计算连续体已被引入作为一种可行的长期经济模型。在这种情况下，在系统中的最佳处理点（边缘，雾，云）利用适当的计算资源的问题变得至关重要。在本文中，我们通过引入互惠计算的新概念来解决这个问题。具体来说，我们不只是考虑参与者之间的交互，而是考虑边缘、雾和云计算提供商与用户建立了一种强制性的共生关系，以满足他们的计算需求。一个网络经济学为基础的分析，以共同使计算提供商和用户，以最大限度地提高他们的利润和效用，分别介绍。两个定价模型进行了研究，以确定计算提供商的最优价格方面的利润最大化。第一种是自由市场定价模型，它使计算提供商之间能够充分竞争，并被表述为计算提供商之间的非合作博弈。第二个-寡头垄断定价-在云计算提供商（领导者）和边缘，雾计算提供商（追随者）之间引入，并通过Stackelberg游戏进行处理。所提出的方法的性能评估是通过建模和仿真实现的，并强调了互利计算范式内的两种不同的定价模型的权衡。
In the transformative data-driven era, Compute Continuum – consisting of the edge, fog and cloud - has been introduced as a viable long-term economic model. In this setting the problem of utilizing the appropriate compute resources at optimal processing points (edge, fog, cloud) in the system, becomes of paramount importance. In this paper, we address this issue by introducing the novel concept of mutualistic computing. Specifically, we do not simply account for the interactions among the involved actors, but we consider that the edge, fog, and the cloud computing providers establish an obligate symbiosis with the users in order to serve their computing demands. A network economics-based analysis is introduced in order to jointly enable the computing providers and the users to maximize their profit and utility, respectively. Two pricing models are studied to determine the computing providers' optimal prices in terms of maximizing their profit. The first one - free market pricing model - enables the full competition among the computing providers and is formulated as a non-cooperative game among the computing providers. The second one - oligopoly pricing - is introduced among the cloud computing provider (leader) and the edge, fog computing providers (followers), and is treated via a Stackelberg game. The performance evaluation of the proposed approach is achieved via modeling and simulation, and the tradeoffs of the two different pricing models within the mutualistic computing paradigm are highlighted.