Evaluation of Process Arrangement Methods Based on Resource Constraint for IoT System

Abstract

In this paper, we evaluate the computation time taken to create a timely optimized scheduler that needs to adhere to the resource constraint by comparing our proposed Sort and Fit method with trivial methods using Brute-force and 0--1 Knapsack Problem. When unoptimized, the scheduler not only takes a longer duration but also increases the operating cost. Due to the scheduling problem itself is an NP-hard, the simplest solution requires an upgrade to expand the resource constraint. Although there are multiple solutions, it cannot be applied in a real-time system and limited only to a specific problem. The major contribution of this paper is our proposed method can compute in near real-time while the accuracy is comparable to the trivial methods in near real-time. This show that trivial methods are suitable for small scale IoT system while our proposed method is suitable for large scale IoT system.