application level clustering
load balncing gives u allredundancy u need
app level clustering is chaos
java
low performance high complexity see
www.tcl.tk
www.cliki.net
www.aolserver.com
www.squeak.org
www.haskell.org
gforth
a ticketing system that does not key off of server name
like a medical history of a patient having line items below the server tells you all that has been wrong and you dont repeat yourself
insaen to ticket any other way
SAN and any storage over network
do commodity arrays attached by fast local wire and fast backplane instead and software raid so the os can alert you of a disk fail
solaris/centos
archlinux or bsd instead
oracle
postgresql NOT enterprise DB
apache
anything event based or aolserver, don't forget about embedit::webit cherrypy yaws cherokee[a favorite of mine] mongrel2 starman swazoo happstack
ksh/bash/perl/python/ruby
see tcl
clisp
forth
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment