In this repository you can find simple Ignite/Spark benchmarks where both of them trying to build ML model based on the same dataset. (especially preprocessed homecredit dataset). Please, contact to authors to get access to this dataset.