PROMISE: Pricing-driven Resource Allocation Method for Service Infrastructure Deployment — Laboratory Package

Deploying applications across the computing continuum requires selecting infrastructure nodes that jointly satisfy functional and non-functional constraints. As infrastructures grow in scale and heterogeneity, this resource allocation (RA) problem becomes inherently combinatorial and can be viewed as an instance of a configuration problem. Existing solutions, however, rely on ad-hoc formulations that hinder reuse and overlook constraints arising in multi-provider environments. This paper instantiates the Configuration Problems framework for resource allocation by representing its configuration space as an iPricing, a model originally proposed for the Software-as-a-Service domain. This provides the first evidence supporting the hypothesis that a unified formulation would be enough to represent configuration problems across the continuum. Building on this, we present: (i) a pricing-based formulation of RA; (ii) \approachName, which leverages a pricing analysis engine to solve resource allocation in multi-provider environments; (iii) synthetic generation processes for infrastructure topologies and workload demands; and (iv) a benchmark comprising 9,600 precomputed RA scenarios. This repository serves as an artifact accompanying the paper described above. It contains the code, datasets, and instructions necessary to reproduce the experiments and results presented in the paper. For instructions on reproducing the experiments, please refer to the README.md file located in the root directory of the artifact. For a detailed presentation of the experimental results reported in the paper, please refer to the Technical Report located in the root directory of the artifact. Repository: https://anonymous.4open.science/r/resource-allocation-5504/