On Sat, Oct 2, 2010 at 7:57 PM, Offer Baruch <[email protected]> wrote:

> Then we used IBM System Automation to cluster both MQ and MB and dropped the
> NFS solution...

Should not be a surprise I think. Implemting HA on a low level
requires that you do it for all the functions, not just the ones that
your application requires. Doing it at applicaiton level means that
you only cover the functions that you require (but you need to do it
for each application). For example providing a HA data storage is
pretty complicated. If you only need that for your web servers' static
data (that does not change a lot) you may have easier options (like
data duplication or accept a brief outage). The easiest obviously is
the user-level: have the user to agree they will accept an outage for
recovery.

PS System z does an amazing job doing the hard part: provide high
availability at a very low level so that your applications don't need
to worry about it.

| Rob

----------------------------------------------------------------------
For LINUX-390 subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO LINUX-390 or visit
http://www.marist.edu/htbin/wlvindex?LINUX-390
----------------------------------------------------------------------
For more information on Linux on System z, visit
http://wiki.linuxvm.org/

Reply via email to