Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Combining Declarative and Linear Programming for Application Management in the Cloud-Edge Continuum

Jacopo Massa, Stefano Forti, Patrizio Dazzi, Antonio Brogi

TL;DR

Problem: efficient, data-aware placement of multi-component applications in the Cloud-Edge continuum under dynamic infrastructure. Approach: a hybrid EdgeWise framework—first a declarative preprocessing stage (with continuous reasoning in EdgeWiseCR) to filter candidates and reuse placements, then a MILP optimizer to obtain cost-efficient mappings. Contributions: an extended EdgeWiseCR pipeline with continuous reasoning, an open-source implementation, and extensive experiments up to 2048 nodes showing up to 65% faster planning and improved stability with modest cost increases. Significance: enables scalable, adaptive deployment of data-intensive, multi-service applications in real-world, failure-prone Cloud-Edge environments.

Abstract

This work investigates the data-aware multi-service application placement problem in Cloud-Edge settings. We previously introduced EdgeWise, a hybrid approach that combines declarative programming with Mixed-Integer Linear Programming (MILP) to determine optimal placements that minimise operational costs and unnecessary data transfers. The declarative stage pre-processes infrastructure constraints to improve the efficiency of the MILP solver, achieving optimal placements in terms of operational costs, with significantly reduced execution times. In this extended version, we improve the declarative stage with continuous reasoning, presenting EdgeWiseCR, which enables the system to reuse existing placements and reduce unnecessary recomputation and service migrations. In addition, we conducted an expanded experimental evaluation considering multiple applications, diverse network topologies, and large-scale infrastructures with dynamic failures. The results show that EdgeWiseCR achieves up to 65% faster execution compared to EdgeWise, while preserving placement stability under dynamic conditions.

Combining Declarative and Linear Programming for Application Management in the Cloud-Edge Continuum

TL;DR

Problem: efficient, data-aware placement of multi-component applications in the Cloud-Edge continuum under dynamic infrastructure. Approach: a hybrid EdgeWise framework—first a declarative preprocessing stage (with continuous reasoning in EdgeWiseCR) to filter candidates and reuse placements, then a MILP optimizer to obtain cost-efficient mappings. Contributions: an extended EdgeWiseCR pipeline with continuous reasoning, an open-source implementation, and extensive experiments up to 2048 nodes showing up to 65% faster planning and improved stability with modest cost increases. Significance: enables scalable, adaptive deployment of data-intensive, multi-service applications in real-world, failure-prone Cloud-Edge environments.

Abstract

This work investigates the data-aware multi-service application placement problem in Cloud-Edge settings. We previously introduced EdgeWise, a hybrid approach that combines declarative programming with Mixed-Integer Linear Programming (MILP) to determine optimal placements that minimise operational costs and unnecessary data transfers. The declarative stage pre-processes infrastructure constraints to improve the efficiency of the MILP solver, achieving optimal placements in terms of operational costs, with significantly reduced execution times. In this extended version, we improve the declarative stage with continuous reasoning, presenting EdgeWiseCR, which enables the system to reuse existing placements and reduce unnecessary recomputation and service migrations. In addition, we conducted an expanded experimental evaluation considering multiple applications, diverse network topologies, and large-scale infrastructures with dynamic failures. The results show that EdgeWiseCR achieves up to 65% faster execution compared to EdgeWise, while preserving placement stability under dynamic conditions.

Paper Structure

This paper contains 14 sections, 3 equations, 4 figures, 3 tables.

Table of Contents

  1. Introduction
  2. Use Case and Considered Problem
  3. Considered problem
  4. Declarative & MILP formulation
  5. Background: Prolog
  6. Declarative model formulation
  7. MILP prototype
  8. Declarative Preprocessing
  9. Pipeline Overview
  10. Experimental Evaluation
  11. Experimental Setup
  12. Discussion of results
  13. Related Work
  14. Conclusions

Figures (4)

  • Figure 1: Example application.
  • Figure 2: Declarative preprocessing step with continuous reasoning.
  • Figure 3: Overview of the EdgeWise pipeline, consisting of a declarative preprocessing stage and a MILP-based placement search phase. With EdgeWiseCR we introduced continuous reasoning in the first stage, and the reuse of previously found eligible placements.
  • Figure 4: Execution times and deployment costs across the three scenarios for each reasoning strategy. Each row corresponds to one application, and each column to a planning version.