Challenges in Scheduling Aggregation in Cyberphysical Information Processing Systems

Data aggregation is an important element in information processing systems, including MapReduce clusters and cyber physical networks. Unlike simple sensor networks, all the data in information processing systems must be eventually aggregated. Our goal is to lower overall latency in these systems by intelligently scheduling aggregation on intermediate routing nodes. In order to understand the potential challenges associated with constructing a distributed scheduler that minimizes latency, we developed a simple model of wireless information processing systems and simulation of our model. Unlike previous models, our model explicitly takes into account link latency and computation time. Our model also considers heterogeneous computing capabilities. We tested the latency while randomly assigning aggregation computation to nodes in the network. Preliminary results indicate that in cases where the computation time is greater than transmission time, in-network aggregation can have a large effect (reducing latency by 50% or more). However, naive scheduling can have a detrimental effect. Specifically, when the root node (a.k.a the base station) is faster than the other nodes, the latency can increase with increased coverage, and these effects vary with the number of nodes present.