Check the preview of 2nd version of this platform being developed by the open MLCommons taskforce on automation and reproducibility as a free, open-source and technology-agnostic on-prem platform.
Pamela project - slambench crowdtuning scenario
Demo of the multi-objective autotuning of slambench via CK (can be any objective, i.e. execution time, energy, code size, cost, precision, reliability). Black dot - GCC with -O3. Green dot - LLVM with -O3
Select subgraph: 
Graph:
Graph params (to customize/reproduce):
[ ]  [ ]    Auto-replot graph: , seconds:
Reproducing graph:
Experiment entries: pamela-autotune-flags-slambench-linux-i10
pamela-autotune-flags-slambench-linux-best
pamela-autotune-flags-slambench-linux-i10-pareto
pamela-autotune-flags-slambench-linux-best-llvm
Scripts to rebuild: ck find script:bfd7ae44dda619dc
View in CK viewer
Replay graph from CMD: ck replay graph:pamela-autotune-flags-slambench-linux id=2d_points_time_vs_size_with_pareto
Obtain shared CK repo with all artifacts: ck pull repo:reproduce-pamela-project
Original experiment table: Download in JSON;   Download in CSV
Embedd original image into interactive report/paper: <img src="https://cKnowledge.io/c/2d41f89bcf32d4d4/3645af97abb448ab/?action=download&filename=2d_points_time_vs_size_with_pareto.png" width="700" height="400" >