Hi Anthony, As you say, the context does soften that sentence.
However, the existing wording is not only wrong, it also gave someone on this list some unnecessary cause for concern. Regards, D On Jan 26, 5:29 pm, Anthony <[email protected]> wrote: > On Wednesday, January 26, 2011 10:45:45 AM UTC-5, villas wrote: > > > Massimo, > > > > "Even if there are multiple web servers, there must be one, and only > > > one, database server. " > > > This does appear to negate one of Web2py's best features, concurrent > > access to multiple DB servers. How about changing the book to read as > > follows (incorporating Jonathan's comment)?? > > > "Even if there are multiple web servers, there must still be a single > > *logical* database server that presents a single, consistent view of > > the data". > > > -D > > Yes, the statement in the book is clearly a general statement about the > scalability challenges of RDBMS's with ACID transactions, not a limitation > specific to web2py. There's a hint at possible solutions in the very next > paragraph: > > "In the rest of the chapter, we consider various recipes that may provide an > improvement over *this naive approach*..." (emphasis added). > > And things are really cleared up further down the page in the "Using > Replicated Databases" > section:http://web2py.com/book/default/chapter/11#Using-Replicated-Databases
