Intelligent Sampling is a novel technique for improving sample quality in large-scale and heterogeneous infrastructures.
We provide a method, based on the Bayes Theorem to sample the infrastructure nodes more accurately, considering the specific incoming workload and our heterogeneous infrastructure model.
We leverage the Alibaba PAI dataset to show that our approach is 2.5x times more accurate compared with other state-of-the-art sampling mechanisms while retaining comparable performance and scalability.