Achieving Observability on Fog Computing with the use of open-source tools

TL;DR

Addresses the challenge of enabling observability in Fog Computing under resource constraints by formalizing Fog Observability and the Fog Observability Data Life Cycle (ODLC), and by evaluating open-source tooling on a real smart-city testbed. The approach centers on a quantitative overhead model, $Outcome = \frac{Observability}{Overhead}$, and a weighted variant $Outcome = \frac{W_{Met}}{Over_{Met}} + \frac{W_{Log}}{Over_{Log}} + \frac{W_{Tra}}{Over_{Tra}} + \frac{X(ID)}{Over_{X(ID)}}$, to guide adaptive data collection. The contributions are a formal definition of fog observability, the ODLC framework, and an empirical evaluation demonstrating that observability can be delivered at the edge with manageable overhead by reducing data volume and carefully configuring the data-path. The findings provide practical guidance for SLA-oriented, edge-centric decision making in IoT deployments and highlight the need for orchestration to balance data richness against resource usage.

Abstract

Fog computing can provide computational resources and low-latency communication at the network edge. But with it comes uncertainties that must be managed in order to guarantee Service Level Agreements. Service observability can help the environment better deal with uncertainties, delivering relevant and up-to-date information in a timely manner to support decision making. Observability is considered a superset of monitoring since it uses not only performance metrics, but also other instrumentation domains such as logs and traces. However, as Fog Computing is typically characterised by resource-constrained nodes and network uncertainties, increasing observability in fog can be risky due to the additional load injected into a restricted environment. There is no work in the literature that evaluated fog observability. In this paper, we first outline the challenges of achieving observability in a Fog environment, based on which we present a formal definition of fog observability. Subsequently, a real-world Fog Computing testbed running a smart city use case is deployed, and an empirical evaluation of fog observability using open-source tools is presented. The results show that under certain conditions, it is viable to provide observability in a Fog Computing environment using open-source tools, although it is necessary to control the overhead modifying their default configuration according to the application characteristics.

Figures (6)

• Figure 1: Overview of the Fog Computing Architecture.
• Figure 2: Fog observability’s data life cycle.
• Figure 3: Fog Experimental Testbed.
• Figure 4: Assessing overhead of observability tools on: IoT devices (a) and (b); Fog Nodes (c) and (d).
• Figure 5: Observability data visualisation on Fog Layer: Metrics (a), Logs (b), and Traces (c) and (d).