Efficient large-scale configuration via integer linear programming

Abstract

Configuration of large-scale applications in an engineering context requires a modeling environment that allows the design engineer to draft the configuration problem in a natural way and efficient methods that can process the modeled setting and scale with the number of components. Existing configuration methods in artificial intelligence typically perform quite well in certain subareas but are hard to use for general-purpose modeling without mathematical or logics background (the socalled knowledge acquisition bottleneck) and/or have scalability issues. As a remedy to this important issue both in theory and in practical applications, we use a standard modeling environment like the Unified Modeling Language that has been proposed by the configuration community as a suitable object-oriented formalism for configuration problems. We provide a translation of key concepts of class diagrams to inequalities and identify relevant configuration aspects and show how they are treated as an integer linear program. Solving an integer linear program can be done efficiently, and integer linear programming scales well to large configurations consisting of several thousands components and interactions. We conduct an empirical study in the context of package management for operating systems and for the Linux kernel configuration. We evaluate our methodology by a benchmark and obtain convincing results in support for using integer linear programming for configuration applications of realistic size and complexity.

A Formal Approach to Microservice Architecture Deployment

“…Those tools support the specification of deployment plans but they do not support the automatic distribution of software instances over the available machines. The proposals closest to ours are those by Feinerer [31] and by Fischer at al. [32].…”

On the Integration of Automatic Deployment into the ABS Modeling Language

“…We defined the capacity of each resource for several AWS instance types in the locations file. 8 The price used in the cost attribute of each AWS instance type concerns on-demand instances in the US East region running Linux.…”

“…Two recent efforts, Feinerer's work on UML [8] and Engage [10], are more similar to our approach as they both rely on a solver to plan deployments. Feinerer's work is based on the UML component model, which includes conflicts and dependencies, but lacks the aspects concerning virtual machines and deployment.…”

Automated synthesis and deployment of cloud applications

“…These constraints distinguish our approach from others [13,12,14], they are necessary due to the flexible dependencies we have on ports. Constraints are constructed using auxiliary variables B(p, tr, tp) for the number of bindings on port p between requesting instances of type tr and providing instances of type tp.…”

“…Two recent efforts, Feinerer's work on UML [12] and Engage [14], are more similar to our approach as they both rely on a solver to plan deployments. Feinerer's work is based on the UML component model, which includes conflicts and dependencies with capacity constraints, but uses dependencies only between components, which greatly restricts the expressiveness of the model (choices are not possible).…”