Domas tells me that that extrapolation leads to some
concern that we
might approach our hardware limits on the s1 cluster. The sums need to
be done and the hardware ordered, if necessary.
flaggedrevs adds 30% database size tax on dewiki.
That would mean that we'd need new-class machines (1TB of fast raid10
storage for upcoming year, may need even bigger storage eventually, if
we continue to double database size yearly, 64-128G of RAM) for enwiki
cluster.
I guess that is much cheaper now, than few years ago. Although, such
data expansion postpones our ability to use enterprise flash at
reasonable price level :)
Alternative is of course analyzing where is all that FR bloat =)
Domas