C
ChUIMonster
Guest
To clarify my comment above -- when I say 50 to 100 records I mean 50 to 100 per timestamp. There are hundreds of millions of records in total -- so the difference is magnified considerably across that dataset. The inefficiencies of the normalized approach were not immediately apparent when there were only millions of records. When there got to be tens of millions it started to become apparent that this was maybe not the best approach...
Continue reading...
Continue reading...