At one of our customers with a PI 2010 installation, performance of aggregate and piplot queries over longer time periods (eg. 6 months) is sometimes (when tag is not in cache?) not good. At the same time we observe that disk IO levels out at 1 MB/s which seems low. The system and disk are capable to run at much higher IO rates and the network does not seem to be the bottleneck. Any hints on increasing performance for such queries on PI 2010? Would performance be better on the most recent version of PI?