Unni Pillai

Hi,

I am also looking for TOGAF 9 Certification . Is the training mandatory or self study is permitted. My undersating for TOGAF 8 is that the training is mandatory. did they changed in 9?

Thanks,
Unni

Thanks for the quick reply !

@Campbell Ritchie The reason why I need to do is something like, we have some APIs which are going to to be rewritten soon.. but will be used by other developers . so we have to warn them saying that this class will be changed soon.. I know, this is kind of wired but.....

Hi,

I have to add @deprecated tag to about 300 java files . Is there any plugins available to do that at once? or I have to open each one of them and do that :-)

Any ideas appreciated.

Thanks

Servlet container is part of application server. For example Jboss application sever uses tomcat servet container.

Hi,

To me that is only a guide line.In my opinion, you should concentrate on the assignment at the best rather than worried about the date. do your best and submit.

Never think that, It is something like, you can submit version 1 now and Oracle will evaluate and say 'correct this part' then you are done.

Oracle changed the grading policy after I submit my assignment, they introduced part minimums, what can we do ?

Bottom line is complete the assignment with the mind set of want to will at the first go. If you can wrap things up by August 27th, that is great but even if you are not you are not loosing anything if you win at first go.


Unni

I am not getting what is your point.

If your requirement says "Java Based System" then you could have a WS call or EJB call or what ever. If your requirements specify it is a webservice then it is better to keep that way.

Why you want to change WS to a EJB call? I tried to design everything open standard as possible so that we can explain future proofing( eg. mobile application needs to check inventory).

Regarding the security, the assumptions is valid.

Unni

Hi Nilessh,


I did it in the deployment diagram html , below the diagram as "Hardware Selection" . I think that is the best place to catch the evaluators attention being in the correct context.

For part 3, I haven't mentioned anything very particular about the calculations or type of hardware, but For NFR related questions I used this knowledge to prove that I considered all of them from the hardware point of view.

The point is , we have to show that, as an architect we are aware of capacity planning and used that knowledge in the solution.

Unni

I found the following post really helpful in understanding about the capacity planning. I didn't show any big calculations but justified the hardware specs based on my assignment

Read This


Hope this helps !

Unni

My opinion is, you should assume 99.99% 24/7 availability if not specified otherwise. So it is better to show fail over mechanisms with stand by servers and all.

Cost and complexity is not a constrain for the assignment. In real life things are different.

So it is safe to have a grand deployment diagram which suits to your assignment, remember you can't argue with the evaluator and prove that your laptop server is good enough to handle 200 users why to waste money in servers and all ( even though it might be true)

Again,this was my thought process when I did it.

Unni

Hello Carlos,

Here is what I did. In my case, I took the user management of an online shopping system as already existing or out of scope for the SuD.

If you think, it is very big assumption to make but it was fair to do based on my assignment. So the bottom line is you can assume any part of the assignment out of scope if it is not the crux of the assignment.

Even though I took user management as out of scope, I have shown it in the component diagrams,sequence diagrams etc as a system(no internal details)

Unni

I also did the same way.

Mittal,

You are correct. You can define it as a component without giving any internal details. But clearly write that in the assumptions, that is very important.

Unni

As long as you document the assumptions clearly. it is Ok.

It is used as a label for different system resources.

For example : Profile A (CPU - 2 GHz, RAM - 4GB,Storage -250GB)
Profile B (CPU - 1GHz RAM -2GB ,Storage -100GB)

I used same profile for all my active and stand by severs.

I used EJB Timer for that purpose in my assignment. In real life I am using Quartz , but as Jeanne pointed out its always better to stick to JEE stuff in the exam

Hi Kumar,


I used if-else condition in the sequence diagrams in a limited way. As long as the diagram is not getting complicated it is ok. If you have too many condition check. then think about splitting the diagrams.I have done that for authentication in my case.

Regarding your Q2, I completely omitted , 'what kind of component' information from my sequence diagrams, according to me, it is for explaining the message flow. In real life also we don't show component details in sequence diagrams.

Unni