It's all about the answers!

Ask a question

Alternative process scenarios for parallel development


Andrés Guerrero (20613218) | asked Mar 24 '09, 7:51 a.m.
Hi guys,
I'm trying to solve the next scenario:

In one Project Area there will be 2 kinds of development:

Main development: "big" developments with its own "budget" and its own phases.

Maintenance development: Little enhancements, bug resolution, maintenance/administrative tasks. These activities will be planned in years.


There will be approximately 100 - 120 Main developments per year on this project area and maybe 40 will be concurrent.

I can create a dev line for maintenance activities and create a phase per year.

But my thought is that creating an independent dev line for each main development has no sense (to much dev lines!!)

I don't know if this alternative could have sense (I know it looks strange). I created these process iterations trying to show a possible solution:

--Main developments (dev_line)

--Dev01 (iteration)
--M1 (iteration)
--Dev02 (iteration)
--M1 (iteration)
--DevN (iteration)
--M1

--Maintenance (dev_line)
--2008 (iteration)
--2009 (iteration)
--2010 (iteration)



There are 2 main development lines. One of them is for maintenance activities and the other one for all the main developments. The point here is that the main developments dev line has N iterations representing N big developments (Dev01, Dev02, Dev03,.., Dev110 ) and these developments are executed in parallel. In addition I can say that these developments are owned by the same set of teams.

On the other hand we are studying the possibility of using releases as the version in which a concrete requirement will be deployed (different case for bugs where the release shown makes reference to the version where the bug is found ). So Iteration is the WHEN will be developed and the version (release) WHERE a feature will be delivered.

I know that dev line has an implicit sense of time flow. There is an active iteration at a time.
In addition to the evident philosophical problem, Is there any other technical problem with this solution?

Am I right thinking that dev lines are harder to maintain and have a higher cost than iterations?.
Any feedback will be appreciated
Thanks a lot

Best regards

Andres

One answer



permanent link
Jared Burns (4.5k29) | answered Mar 27 '09, 3:16 p.m.
FORUM ADMINISTRATOR / FORUM MODERATOR / JAZZ DEVELOPER
On Tue, 24 Mar 2009 11:59:47 +0000, jaguerrero wrote:
Am I right thinking that dev lines are harder to maintain and have a
higher cost than iterations?.
Any feedback will be appreciated
Thanks a lot

Best regards

Andres


The biggest overhead cost of multiple development lines is the fact that
each development line has a separate hierarchy of team areas. For
situations where you have an identical structure of 10 teams working on 2
concurrent development lines, this means you'd need 20 team areas which
can be inconvenient.

However, in your case I'm guessing this isn't a problem. You said that
you've got 40 separate schedules going concurrently, which can't possibly
be handled by the same teams. So in your case, I would guess that the
separate team area hierarchies would be a feature. For each of your
concurrent development lines, you'll have an accurate representation of
which teams are actually doing the work.

--
Jared Burns
Jazz Process Team

Your answer


Register or to post your answer.


Dashboards and work items are no longer publicly available, so some links may be invalid. We now provide similar information through other means. Learn more here.