PiBench Online is an interactive system for benchmarking PM indexes in a fair and reproducible manner, it is based on PiBench, the first unified, highly customizable benchmarking framework for PM indexes.
PiBench Online enables push-button evaluation of PM indexes: the user can upload a shared library that implements the index, set parameters to run customized benchmarks, and analyze results interactively, all through an easy-to-use web interface.
We have made PiBench Online available at http://pibench.org and hope it can promote fair comparison and reproducibility in future PM index research.
PiBench Online is built by: Xiangpeng Hao, Tianzheng Wang ( Systems Group, Simon Fraser University), Lucas Lersch (TU Dresden & SAP SE) and Ismail Oukid (Snowflake Computing)
Read our paper: Evaluating Persistent Memory Range Indexes
Contribute to PiBench: PiBench GitHub
Discussion: Google Group
Request PiBench resources: support@pibench.org