[Seaside] Seaside Performance
dsalama at user.net
Sat Mar 26 06:44:03 CET 2005
I was testing some file I/O reading when I noticed something peculiar.
I posted the following on the Squeak mailing list
090187.html). Because of this finding, I am wondering how well will a
Squeak box perform in a production environment.
I'm working on duplicating a CRM application I developed using Ruby On
Rails with MySQL. This application currently runs on a single server
with dual 650Mhz CPUs and 2GB of RAM. Currently, my client has over 1
million customer records and each record has at least one order in the
system. They handle approximately 10000 application "transactions" per
day (a transaction could be something as simple as looking up a
customer profile, to something more complex such as placing and order,
while interfacing with USPS using SOAP and the corresponding financial
institutions using XML to verify payment information) with
approximately 50 concurrent users.
My idea was to replicate that same environment with
Squeak/Seaside/GOODS and, if everything worked well, eventually migrate
to that environment. After seeing what I saw today regarding the post I
mentioned above, I'm concerned this environment may not necessarily
scale as needed for this application. Would anyone care to comment? Are
there ways of clustering Squeak/Seaside? I don't necessarily think
GOODS will be a bottleneck, unless accessing GOODS will create some
sort of I/O bottleneck.
More information about the Seaside