Recently at Scalathon I have been made aware of this article, detailing a data structure for high-throughput data processing. I could not help thinking that the ideas expressed in that article are sound very familiar: while doing my PhD research at the COMPASS experiment at CERN, I implemented an online event filter in 2004 which utilized the same data structure for the central stream buffer. I shortly described it in my PhD thesis (page 38ff) and an IEEE conference record for NSS/MIC 2007. We named the application “Cinderella” after the fairy-tale, as its purpose was to sort the bad (i.e. uninteresting) events from the stream of good ones before it hit the disk buffers.