This paper describes Nancy, a Network Calculus (NC) library that allows users to perform complex min-plus and max-plus algebra operations efficiently. To the best of our knowledge, Nancy is the only open-source library that implements operators working on arbitrary piecewise-linear functions (as opposed to only concave/convex ones), as well as to implement some of them (e.g. sub-additive closure and function composition). Nancy allows researchers to compute NC results using a straightforward syntax, which matches the algebraic one. Moreover, it is designed having computational efficiency in mind: it exploits clever data representation, it uses inheritance to allow for faster algorithms when they are available (e.g., for specific subclasses of functions), and it is natively parallel, thus reaping the benefit of multicore hardware. This makes it usable to solve NC problems which were previously considered beyond the realm of tractable.
翻译:本文描述了南希,这是一个网络计算库(NC)图书馆,它使用户能够高效率地进行复杂的微增和最大增代数操作。据我们所知,南希是唯一一个开放源码图书馆,它负责操作操作任意的细线功能(而不是仅仅连接/凝固功能),以及执行其中的一些功能(例如子相加关闭和功能构成)。南希允许研究人员使用直截了当的语法计算NC结果,该语法与代数法相匹配。此外,它设计时还考虑到计算效率:它利用智能数据代表,利用继承来允许在具备这些功能时采用更快的算法(例如,对于特定的子功能类别),它是本地平行的,从而获得多芯硬件的好处。这使得它能够用来解决先前被认为超出可引力范围的NC问题。