The increase in satellite launches into Earth's orbit in recent years has generated a huge amount of remote sensing data. These data, in the form of time series, have been used in automated classification approaches, generating land-use and land-cover (LULC) products for different landscapes around the world. Dynamic Time Warping (DTW) is a well-known computational method used to measure the similarity between time series. Tt has been used in many algorithms for remote sensing time series analysis. These DTW-based algorithms are capable of generating similarity measures between time series and patterns. These measures can be used as meta-features to increase the accuracy results of classification models. However, DTW-based algorithms require a lot of computational resources and have a high execution time, which makes them difficult to use in large volumes of data. This article presents a parallel and fully scalable solution to optimize the construction of meta-features through remote sensing time series (RSTS). In addition, results of the application of the generated meta-features in the training and evaluation of classification models using Random Forest are presented. The results show that the proposed approaches have led to improvements in execution time and accuracy when compared to traditional strategies.