New on the Blue Parabola blog today, there's an article from Keith Casey trying to set the record straight on some recent benchmarks put together by Zend and Acquia showing the performance of Drupal.
While I have not attempted to duplicate or validate any of their individual numbers or conclusion as a whole, I have been a Drupal user for well over five years and have launched 30+ sites on it.
With this experience to back him up, he points out a few things about the report that either needed work or were misleading in their results. This includes a limitation on the web server configurations they covered, the statistics on Windows performance (might turn people off to using it on this platform) and the emphasis being put in some of the wrong places - optimizing PHP versus Drupal.
He recommends that, like all studies and whitepapers like this, you take the results as only guidelines and examples, not as absolute facts. There's always going to be differences in hardware, software and configuration so what gave the big numbers and results for them might not work for you.