WhatIsZopeScalability

ZOPE has taken measures to be run on minimalist hardware. This self imposed constraint has turned into a significant advantage when it comes to scalability on single as well as multiple servers. With a single modest server, like zope.org you can handle many concurrent requests. If it needs to be upgraded the usual solution will be add memory or get a new processor. This is the way of scaling vertically, but the most interesting way is scaling horizontally using many comodity servers with a common Storage - the paradigm behind ZEO. This way if you need to handle another 100 concurrent requests, you can just snap in a 1U cobalt box, rocket logix, or the like!!

• Singular ZOPE (single server)
  • Python Interpreter - what is the footprint(?)
  • SMP machines - dont know what to say here :(
• Distributed ZOPE (multiple servers)
  • ZEO Fact Sheet is available here
  • ZEO appears to be to fairly solid on *nixes, but may have some minor problems on NT.

My use of ZOPE has been a behind the scenes approach where traffic does not actual occur on ZOPE itself. I think lots of people will find its powerful OOP web development environment as enough of a winner for it to be used not necessarily as the production environment but as a supplement to their environments. My case is political in nature. I feel dirty speaking management talk, but I like it.