chaoshunh / amp Goto Github PK
View Code? Open in Web Editor NEWThis project forked from blindsub/amp
We propose a deep learning based cloud workload prediction algorithm with Adaptive Pattern Mining (AMP). First, a two-step, multi-models based method is proposed to capture the ”low frequency” and ”high frequency” characteristics of highly-variable workloads simultaneously. Next, we design a DNN based classification algorithm to automatically identify the different workload patterns for a given workload and use the corresponding model to predict the future workload.