Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Privacy-Preserving Coding Schemes for Multi-Access Distributed Computing Models

Shanuja Sasi

TL;DR

This work introduces privacy constraints into the multi-access distributed computing (MADC) framework and develops private coded schemes for two distinct connectivity patterns. By constructing extended placement delivery arrays (PDAs) and leveraging PDA-based coding, the authors achieve private coded schemes with computation load $r=1$ and explicit communication loads $L_p$ that depend on system parameters. The two main models, α-connect and α-cyclic PPMA-MADC, are supported by algorithmic constructions (Algorithms 1 and 3) and two corresponding constructions (Construction 1 and Construction 2), with detailed proofs of correctness and privacy. The proposed schemes are particularly relevant for edge and multi-tenant environments, where preserving the privacy of each reducer's assigned function index is as important as minimizing data replication and communication overhead.

Abstract

Distributed computing frameworks such as MapReduce have become essential for large-scale data processing by decomposing tasks across multiple nodes. The multi-access distributed computing (MADC) model further advances this paradigm by decoupling mapper and reducer roles: dedicated mapper nodes store data and compute intermediate values, while reducer nodes are connected to multiple mappers and aggregate results to compute final outputs. This separation reduces communication bottlenecks without requiring file replication. In this paper, we introduce privacy constraints into MADC and develop private coded schemes for two specific connectivity models. We construct new families of extended placement delivery arrays and derive corresponding coding schemes that guarantee privacy of each reducer's assigned function.

Privacy-Preserving Coding Schemes for Multi-Access Distributed Computing Models

TL;DR

This work introduces privacy constraints into the multi-access distributed computing (MADC) framework and develops private coded schemes for two distinct connectivity patterns. By constructing extended placement delivery arrays (PDAs) and leveraging PDA-based coding, the authors achieve private coded schemes with computation load and explicit communication loads that depend on system parameters. The two main models, α-connect and α-cyclic PPMA-MADC, are supported by algorithmic constructions (Algorithms 1 and 3) and two corresponding constructions (Construction 1 and Construction 2), with detailed proofs of correctness and privacy. The proposed schemes are particularly relevant for edge and multi-tenant environments, where preserving the privacy of each reducer's assigned function index is as important as minimizing data replication and communication overhead.

Abstract

Distributed computing frameworks such as MapReduce have become essential for large-scale data processing by decomposing tasks across multiple nodes. The multi-access distributed computing (MADC) model further advances this paradigm by decoupling mapper and reducer roles: dedicated mapper nodes store data and compute intermediate values, while reducer nodes are connected to multiple mappers and aggregate results to compute final outputs. This separation reduces communication bottlenecks without requiring file replication. In this paper, we introduce privacy constraints into MADC and develop private coded schemes for two specific connectivity models. We construct new families of extended placement delivery arrays and derive corresponding coding schemes that guarantee privacy of each reducer's assigned function.
Paper Structure (21 sections, 43 equations, 1 figure, 3 algorithms)

This paper contains 21 sections, 43 equations, 1 figure, 3 algorithms.

Table of Contents

  1. Introduction
  2. Problem Definition
  3. Placement Delivery Array Background
  4. Private Coded MADC Model
  5. Coding Scheme for Private $\alpha$-connect PPMA-MADC model
  6. Overview of Algorithm 1 and Construction \ref{['CON1']}
  7. Private coding scheme
  8. Coding Scheme for Private $\alpha$-cyclic PPMA-MADC model
  9. Construction of $(1,2)$-regular PDAs using Algorithm 3
  10. Private Coding Scheme for $\alpha$-cyclic PPMA-MADC model
  11. Proof of Algorithm 1
  12. Proof of Construction 1
  13. Proof of Theorem \ref{['thm1']}
  14. Map Phase
  15. Shuffle Phase
  16. ...and 6 more sections

Figures (1)

  • Figure 1: An MADC model with $\Lambda$ mapper nodes and $K$ reducer nodes.

Theorems & Definitions (8)

  • Definition 1: Computation Load
  • Definition 2: Communication Load
  • Definition 3
  • Definition 4
  • Definition 5
  • Example 1
  • Example 2
  • Example 3