With an operating system chosen, we can move on to learning about and configuring the first two layers of the stack, Zope itself and the ZEO database server.
A typical desktop installation of Plone involves a single Zope process, which reads from and writes to a Data.fs
file directly. ZEO, Zope's specialized database server, takes over that responsibility, and one or more copies of Zope (called Zope instances) communicate with it via a socket. This setup has several advantages:
Scalability. ZEO's main benefit is the ability to run more than one Zope instance at a time on a single data set. Python takes poor advantage of multiple or multi-core processors, and their ubiquity in today's hardware makes this a prime optimization even for single-server deployments. A good rule of thumb is one Zope instance for every logical processor; for example, if you have a single dual-core processor, run two Zopes. Don't worry about leaving a core free for ZEO or other parts of the stack; Zope...