I see you implemented the DPPG algorithm, that's nice.
Have you checked out QTune yet? In Guoliang Li et. al.'s paper "QTune: A Query-Aware Database Tuning System with Deep Reinforcement Learning - ), they suggest an algorithm they call dual-state DPPG, which they show to have further latency and throughput improvements over CDBTune (DPPG).
Have you thought about implementing this algorithm?
Hey guys,
I see you implemented the DPPG algorithm, that's nice. Have you checked out QTune yet? In Guoliang Li et. al.'s paper "QTune: A Query-Aware Database Tuning System with Deep Reinforcement Learning - ), they suggest an algorithm they call dual-state DPPG, which they show to have further latency and throughput improvements over CDBTune (DPPG).
Have you thought about implementing this algorithm?
Link to the paper: https://doi.org/10.14778/3352063.3352129