This is source code for conference paper "A Measurable Framework for Run time Data Sampling in Large scale Datacenter". Old_version is the original code using 'average trick' without code refactoring.
Processed dataset for experiment could be download at https://drive.google.com/open?id=1Lr31aTbA1Xt2NuDz5Cx7WrEHzuh8kkta. Original dataset could be download at https://github.com/alibaba/clusterdata. It's containers' runtime data of alibaba_trace_2017, and we sorted the table by timestamp and ID.