I recently purchased a copy of "The Benchmark Handbook", a book from
the early 1990s that was edited by Jim Gray. It features analysis of
the Wisconsin Benchmark in chapter 3 -- that's a single client
benchmark that famously showed real limitations in the optimizers that
were current in the early to mid 1980s. The book describes various
limitations of Wisconsin as a general purpose benchmark, but it's
still interesting in other ways, then and now. The book goes on to say
that it is still often used in regression tests.
I see that we even had a full copy of the benchmark until it was torn
out by commit a05a4b47 in 2009. I don't think that anybody will be
interested in the Benchmark itself, but the design of the benchmark
may provide useful context. I could imagine somebody with an interest
in the optimizer finding the book useful. I paid about $5 for a second
hand copy of the first edition, so it isn't a hard purchase for me to
justify.
--
Peter Geoghegan
