LEAF: Simulating Large Energy-Aware Fog Computing Environments

Abstract

Despite constant improvements in efficiency, today's data centers and networks consume enormous amounts of energy and this demand is expected to rise even further. An important research question is whether and how fog computing can curb this trend. As real-life deployments of fog infrastructure are still rare, a significant part of research relies on simulations. However, existing power models usually only target particular components such as compute nodes or battery-constrained edge devices.Combining analytical and discrete-event modeling, we develop a holistic but granular energy consumption model that can determine the power usage of compute nodes as well as network traffic and applications over time. Simulations can incorporate thousands of devices that execute complex application graphs on a distributed, heterogeneous, and resource-constrained infrastructure. We evaluated our publicly available prototype LEAF within a smart city traffic scenario, demonstrating that it enables research on energy-conserving fog computing architectures and can be used to assess dynamic task placement strategies and other energy-saving mechanisms.

Simulation Tools for Fog Computing: A Comparative Analysis

“…There are many available simulators that can simulate the scenarios of cloud-fog computing environments such as YAFS [ 22 ], LEAF [ 23 ], EdgeCloudSim [ 24 ], MobIoTSim [ 25 ], SimpleIoTSimulator [ 26 ], IBM BlueMix [ 27 ], Google IoT Sim [ 28 ], iFogSim [ 29 ], Cooja [ 30 ], FogTorch [ 31 ], RECAP [ 32 ], EmuFog [ 33 ]. Most of the available simulators are similar in their functionalities, programming language, or architecture.…”

“…The most recent simulator of this survey is LEAF and is introduced in 2022 [ 23 ]. LEAF can be used for modeling energy-aware scenarios in FC environments.…”

Towards Energy Consumption and Carbon Footprint Testing for AI-driven IoT Services

“…Others, like iFogSim [38] and PureEdgeSim [39], can simulate power utilization, but only for system entities such as cloud and fog nodes (iFogSim) or edge devices and their local network (PureEdgeSim). On the other hand, LEAF [40] is a simulator that primarily focuses on the simulation of power consumption of complete edge and fog systems. While such fog and edge simulators can provide valuable insights by allowing specific scenarios to be assessed, modeling actual software using these simulators is often difficult and requires careful configuration of simulation parameters.…”

Let's wait awhile

“…Since openly available cloud computing data sets that contain information about the delaytolerance of workloads are not available, we created two scenarios ourselves, featuring (1) short-running, periodically scheduled jobs, and (2) long-running machine learning trainings based on the StyleGAN2-ADA [28] paper. The experiments are simulated using LEAF [62], an IT infrastructure simulator that enables high-level modeling of energy consumption. The experimental setup comprises a single node, representing a data center, on which the jobs are scheduled.…”