Novel Constructions for Computation and Communication Trade-offs in Private Coded Distributed Computing

TL;DR

This paper proposes a new private coded distributed computing model that balances computation and communication loads while maintaining privacy, using an extended PDA framework to optimize trade-offs.

Contribution

It introduces an extended PDA framework for private distributed computing, enabling analysis and optimization of computation-communication trade-offs under privacy constraints.

Findings

Privacy increases communication overhead but can be mitigated.

Extended PDAs effectively characterize achievable loads.

Optimized strategies improve trade-offs in private distributed computing.

Abstract

Distributed computing enables scalable machine learning by distributing tasks across multiple nodes, but ensuring privacy in such systems remains a challenge. This paper introduces a novel private coded distributed computing model that integrates privacy constraints to keep task assignments hidden. By leveraging placement delivery arrays (PDAs), we design an extended PDA framework to characterize achievable computation and communication loads under privacy constraints. By constructing two classes of extended PDAs, we explore the trade-offs between computation and communication, showing that although privacy increases communication overhead, it can be significantly alleviated through optimized PDA-based coded strategies.