Software implementation of the Analog Ensemble technique as a way to generate probabilistic forecasts. Provision of several sample research cases, results, and implementation information. Probabilistic forecasts have gained increasing significance in model simulations and economic studies because they estimate uncertainty by providing a distribution of the potential future states of the atmosphere. Probability information is typically derived from an ensemble of deterministic predictions where each prediction corresponds to a single model run. However, numerical models are generally computationally expensive and therefore multiple model simulation raises the problem of computational costs.This tutorial introduces Analog Ensemble (AnEn), which is a technique to estimate the probability distribution using data from observations, a historical repository of corresponding predictions, and only a single model run. Advantages of AnEn include: (1) AnEn easily couples with high-resolution forecast model and does not require initial condition perturbations; (2) AnEn compensates for the model bias by taking historical model errors into consideration which enables the AnEn to generate predictions with improved accuracy; 3) AnEn can predict additional predictand variables that are not present in the weather model as, for example, wind and solar photovoltaic energy production. Accordingly, AnEn can be considered the state-of-the-art technique for generating ensemble forecasts with a computational budget.This tutorial will walk you through the relevant background information about AnEn followed by sample cases showing step-by-step instructions on how to generate probabilistic forecasts. You will use the Ensemble Toolkit (EnTK) to encode the sample cases, executing them with multiple threads and processes on Cheyenne. EnTK is a Python library for developing and executing large-scale ensemble-based workflows, developed by the RADICAL Research Group at Rutgers University. The tutorial will show you how to encode ensemble applications in pipelines, stages, and tasks. This will offer the opportunity to discuss concurrency and parallelism of ensemble execution, showing how they impact on the ensemble time-to-execution and resource sizing. AnEn is co-developed with EnTK and the tutorial will cover the steps that lead to the current implementation, discussing the challenges and opportunities of close collaboration among experts in Earth Science and HPC.