Path Algebra Driven Classification Solution to Realize Performance Oriented VNE Embeddings

The intense diversity of the Next Generation Networking environments creates a dire need for a modular framework that can integrate with metric-based optimization constructs, to tackle on-demand service-centric and user-centric resource allocation scenarios. Our research offers a new evaluation mechanism to successfully replace traditional path ranking and path selection techniques. By applying the fundamental monotonicity and isotonicity properties of the Path Algebra framework, it can always lead to valid and optimal results. Specifically, first we introduce a new performance-oriented monotonic and isotonic synthesized metric. Afterwards, we propose a methodology that analyzes and determines the weighted influence combination depending on the characteristics of the served user-centric application. The chosen suitable weights address performance-oriented mission critical tailored objectives for adaptive optimizations. Its innovative algebraic design allows it to successfully describe and rank candidate paths multidimensionally, whether in legacy or modern architectures. The experimental data show that the proposed framework suggests more suitable paths in each studied scenarios than the compared methodologies, offering customized solutions according to the specific objective.

下一代网络环境的强烈多样性迫切需要一个模块化框架，该框架能够与基于指标的优化结构集成，以应对以服务为中心和以用户为中心的资源分配场景。本研究提出了一种新的评估机制，以成功替代传统的路径排序和路径选择技术。通过应用路径代数框架的基本单调性和等调性属性，它始终能够导向有效且最优的结果。具体而言，首先我们引入了一种以性能为导向的单调性和等调性综合度量指标。随后，我们提出了一种方法论，该方法分析并确定基于所服务以用户为中心的应用特性的加权影响力组合。所选的适宜权重针对性能导向的关键任务定制目标进行适应性优化。其创新的代数设计使得它能够成功多维度地描述和排序候选路径，无论是在传统架构还是现代架构中。实验数据表明，所提出的框架在每个研究场景中都比比较的方法建议了更多合适的路径，并提供了根据特定目标定制的解决方案。