Hyperplot
Lets plot a lot of data!
Tool for efficiently writing partitioned time series data to some file-backend and serving that data in plotly.
Robust to out of order data but not duplicates so best to ingest data on an event.
Open Tasks:
-
Async Read Requests
-
Timestamps should be in nanoseconds -> convert on selection to find partitions to read
-
Better level solver
-
Only redraw graph if partitions have changed
-
Issue where a trace could change subplots when magnitudes are recalculated (maybe okay?)
-
Go back button to go back to a previous view
Open Questions:
- Optimal max points in highest resolution?