Cloud Versus Local Processing in Distributed Networks

TL;DR

This paper presents a method to evaluate the performance trade-offs between local, cloud, and combined processing of divisible data loads within distributed networks, using Amdahl's law as a framework.

Contribution

It introduces a systematic approach to compare local, cloud, and hybrid processing in distributed networks, illustrated through a star network model and various scheduling policies.

Findings

Performance depends on data load partitioning and network configuration.

Hybrid processing can optimize performance for specific applications.

The method applies to mobile computing, cloud services, and signature searching.

Abstract

A method for evaluating the relative performance of local, cloud and combined processing of divisible (i.e. partitionable) data loads is presented. It is shown how to do this in the context of Amdahl's law. A single level (star) network operating under each of three fundamental scheduling policies is used as an example. Applications include mobile computing, cloud computing and signature searching.