#visualization of #time-series #quantiles (on #space-travel #predictions)
on 02015-08-12#Pandas #time-series downsampling uses the how parameter to aggregate points, while upsampling uses the fill_method parameter, generally with "pad".
Discussion of #time-series #databases from 2014. Many people recommend #Postgres for under 10 billion items.
on 02015-08-10“Castra is an on-disk, partitioned, compressed, #column-store with a focus on #time-series applications. It provides efficient columnar range queries.”
on 02015-08-10