Recommended Deployment Strategies for Building FlexStack

Question

I have found the "Recommended deployment strategies for building flexstack"
These are the important items to consider when deploying FlexStack:
• It is recommended that the stack master not contain uplinks if it is possible to distribute the uplinks across other stack member
• When using only two uplinks in a stack of four members, put the uplinks on members that are not directly connected with FlexStack links. Directly connected members are those connected together with FlexStack connections. This is only relevant on FlexStack groups that contain four members.
– This recommendation is purely for stack bandwidth efficiency. Placing the FlexStack uplinks on directly connected members is less efficient on stack bandwidth. In a four-member stack, with members connected 1-2, 2-3, 3-4, 4-1, put the uplinks on members 2 and 4 or 1 and 3.
So For a Stack of 4 WS-C2960S If we use two links to create an etherchannel "Recommended deployment strategies for building flexstack" is to use an interface on Switch 2 and another on  Switch 4 for the etherchannel to Core Switch.
If we want to create an etherchannel of four links, is the recommanded deployment strategies also to use two interfaces on Switch 2 and two interfaces on Switch 4?

Answer 1

Which activities are taking a long time? The ejbc/appc step? or the boot of the
server and deployment of all the beans? Makes a difference, because splitting
up a large number of beans in to multiple JAR files might help with the time it
takes to create each JAR file but wouldn't help with deployment time.
I've seen applications with 300+ entity beans, and while it is a little painful
to run ejbc/appc and deploy the resulting ejb file, it wasn't bad enough to invalidate
the approach or require multiple files. How long are steps taking now? How many
beans, and what hardware are you using for the various steps in the process?
-Greg
http://www.amazon.com/exec/obidos/ASIN/047128128X/
"monica" <[email protected]> wrote:
>
I am working in a medium project that have in average 200 CMP entity
beans. Actually,
we have a mapping between each data base table with a CMP. Every time
that we
develop a new functionality the jar grows up and in fact, the deploying
time grows
up. This is a big problem if we think that only we had develop the 5%
of all project
and it spent much time making building of the jar. I would like to know
if exist
an extrategy to control the growth of the jar, for example do not map
every table
to an entity to make modules, and make build in independent form, one
module for
one group of functionality. The problem is that in this project exist
a group
of entities that are common for every functionality, Country, City, User,
etc..
This is the only strategy in which we could think but we do not have
a clear judgement
to decide which tables must to be mapped, only the ones that are not
catalogs?.
If you can guide me about the better form to organize my project or show
me a
knowledge base with similars problems I would thank you , very much.
Thanks in advance,
Monica A.