This is the eighth progress report from the Wide Finder Project; a quick comment-light aggregation of work that other people have been doing in the space. I’ve managed to get access to an unannounced many-core server and have some preliminary results (summary: Vinoski’s in the lead); I’ll publish those in the, uh, very near future, when things are, uh, less unannounced.
I’ll also make a serious effort to grab all the code I’ve been linking to
here and run it on the T2 and report back, but probably won’t be able to start
next Thursday. By the way, extra credit is given for
solving the whole problem and replicating the all the function of the
original Ruby script; computing the ten most popular fetches by matching a
pattern that includes the
GET in front of the URI.
If you’re working on this and I’ve missed you, shoot me a line.
OK, off to the races.
C · Ilmari Heikkinen has Wide Finder implementations, but seems kind of discouraged and negative about the whole thing.
Python · The Pythonistas have arrived with a mighty slither of scales. Bill de hÓra has a serial implementation, wfinder_serial.py.
Erlang · Pichi follows up on Caoyuan’s work, already reported here, in Binaries really faster than lists.
C# · I also got a damn interesting C# implementation, but it turned out to involve some unreleased infrastructure and its author had to back off. Hey man, I’d love to run that code if you can make it Safe For Work.