Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Exploiting Storage for Computing: Computation Reuse in Collaborative Edge Computing

Xingqiu He, Chaoqun You, Tony Q. S. Quek

TL;DR

Addresses duplicate computations and latency in Collaborative Edge Computing (CEC) by enabling cross-BS computation reuse. The authors formulate a weighted response-time minimization problem that jointly optimizes caching decisions $x^a_{nk}$, cache-searching $y^a_n$, workload distribution $\lambda^a_n$, and CPU allocation $f^a_n$, and solve it via alternating minimization with a bisection-based caching scheme and a projected-gradient-descent scheduler. The problem is NP-hard, but the paper provides polynomial-time approximate algorithms with convergence guarantees. Numerical results show substantial reductions in weighted delay across diverse workloads and network scales, validating the practicality of collaborative caching in edge networks.

Abstract

Collaborative Edge Computing (CEC) is a new edge computing paradigm that enables neighboring edge servers to share computational resources with each other. Although CEC can enhance the utilization of computational resources, it still suffers from resource waste. The primary reason is that end-users from the same area are likely to offload similar tasks to edge servers, thereby leading to duplicate computations. To improve system efficiency, the computation results of previously executed tasks can be cached and then reused by subsequent tasks. However, most existing computation reuse algorithms only consider one edge server, which significantly limits the effectiveness of computation reuse. To address this issue, this paper applies computation reuse in CEC networks to exploit the collaboration among edge servers. We formulate an optimization problem that aims to minimize the overall task response time and decompose it into a caching subproblem and a scheduling subproblem. By analyzing the properties of optimal solutions, we show that the optimal caching decisions can be efficiently searched using the bisection method. For the scheduling subproblem, we utilize projected gradient descent and backtracking to find a local minimum. Numerical results show that our algorithm significantly reduces the response time in various situations.

Exploiting Storage for Computing: Computation Reuse in Collaborative Edge Computing

TL;DR

Addresses duplicate computations and latency in Collaborative Edge Computing (CEC) by enabling cross-BS computation reuse. The authors formulate a weighted response-time minimization problem that jointly optimizes caching decisions , cache-searching , workload distribution , and CPU allocation , and solve it via alternating minimization with a bisection-based caching scheme and a projected-gradient-descent scheduler. The problem is NP-hard, but the paper provides polynomial-time approximate algorithms with convergence guarantees. Numerical results show substantial reductions in weighted delay across diverse workloads and network scales, validating the practicality of collaborative caching in edge networks.

Abstract

Collaborative Edge Computing (CEC) is a new edge computing paradigm that enables neighboring edge servers to share computational resources with each other. Although CEC can enhance the utilization of computational resources, it still suffers from resource waste. The primary reason is that end-users from the same area are likely to offload similar tasks to edge servers, thereby leading to duplicate computations. To improve system efficiency, the computation results of previously executed tasks can be cached and then reused by subsequent tasks. However, most existing computation reuse algorithms only consider one edge server, which significantly limits the effectiveness of computation reuse. To address this issue, this paper applies computation reuse in CEC networks to exploit the collaboration among edge servers. We formulate an optimization problem that aims to minimize the overall task response time and decompose it into a caching subproblem and a scheduling subproblem. By analyzing the properties of optimal solutions, we show that the optimal caching decisions can be efficiently searched using the bisection method. For the scheduling subproblem, we utilize projected gradient descent and backtracking to find a local minimum. Numerical results show that our algorithm significantly reduces the response time in various situations.
Paper Structure (18 sections, 3 theorems, 25 equations, 5 figures, 1 table, 2 algorithms)

This paper contains 18 sections, 3 theorems, 25 equations, 5 figures, 1 table, 2 algorithms.

Table of Contents

  1. Introduction
  2. Related Work
  3. System Model and Problem Formulation
  4. Computation Reuse Model
  5. Task Scheduling Model
  6. Problem Formulation
  7. Algorithm Design and Theoretical Analysis
  8. Problem Relaxation and Decomposition
  9. Caching Subproblem
  10. Scheduling Subproblem
  11. Performance Evaluation
  12. Convergence of the Proposed Algorithm
  13. Impact of Workload
  14. Impact of Network Scale
  15. Conclusions
  16. ...and 3 more sections

Key Result

Theorem 1

The problem lmp is NP-hard even if there is only one BS and one application in the system.

Figures (5)

  • Figure 1: A simple example of the considered system model.
  • Figure 2: Convergence of the proposed algorithm under different learning rates.
  • Figure 3: Impact of workloads.
  • Figure 4: Impact of the number of BSs.
  • Figure 5: Impact of the number of applications.

Theorems & Definitions (6)

  • Theorem 1
  • proof
  • Theorem 2
  • proof
  • Theorem 3
  • proof