Merely the online chapter, I'm afraid. Actually, I've never even heard
of the author before. But as I say, he has some rather intelligent
comments to make about the hit ratio (my only grumble is that, for
Oracle, it would be even more intelligent not to mention it at all!),
and I was overall impressed, albeit on a very thin basis of just one
sample chapter.

Put it this way: unlike certain other authors that could be named, he
*doesn't* say that 'if you can increase the buffer cache hit ratio from
95 to 98%, your database will perform 400% faster' (though I suspect
that your author's reference to seeing "98%" advised in one or more
performance tuning books is a direct reference to that lunatic assertion).

Regards
HJR