Decision Analysis
Sabtu, 16 Maret 2013
0
komentar
Decision Analysis, interesting topic but confusing and difficult one. I started reading some references to know more about DA, but it only helped ot add more grains to my thoughts.
http://en.wikipedia.org/wiki/Decision_analysis
http://home.ubalt.edu/ntsbarsh/opre640a/partix.htm
Let me share my recent experience how I helped myself to reach a solution and share the same to the business for taking easy decision.
Business requirement :
The "PRU web apps" which I would like to discuss is currently runs on IBM WAS 6.x platform with JSF 1.2. The business was forced to migrate the PRU web apps to IBM WAS 7.x platform as the IBM WAS 6.x EOL[End of Life] has reached nearby, which is Sept 13.
Problem :
Team had intiated their migration project and got halted in between as they got some migration issues due to the incompatibility of the JWL libraries[UI Widget libraries] of IBM JSF version of WAS 6.x and WAS7.x and the some of the code they have used in PRU web apps.
Initial Analysis :
Team has contacted IBM Support help, tried all options and finally left to live with the alternative options. Now the business got the red alert and started thinking multiple options, which includes -> Migrating to other JSF implementation, fixing the JWL compatibility issues, any hybrid options or even thinking little more futuristic and migrated to WAS 8.x with MyFaces2.0 which is the recent IBM platform and avoid another probable migration from 7.x etc etc.
Business has their own limitations : They cant use any other JSF implementations other than IBM specific and the PRU Web cant be deployed any other container other than IBM specific Application Servers.
What would be the best option or feasible option or shall we move little cautious way or phased migration.
Solution :
More than the solution, I would like to discuss on how we have arrived on a suitable solution or the best solution approach.
Since the major issue is the compatibility issue of IBM - JWL Widget library with update versions, we have intiated a prototype of replacing JWL libraries with Tomahawk libraries[MyFaces], which is a hybrid approach, and got it working reconfirming our initial analysis. Based on the prototype and long digging of various discussion forums/ IBM support, we have concluded to go for MyFaces 1.2 with WAS 7.x platform.
Now we have created a beautiful diagram of next solution approach. Please find the same below.
The Y-Axis says the future IBM WAS versions, and Y-Axis says the next MyFaces versions. The digram says, once we have reached the immediate solution, we should be able to move to any other combination without much effort.
We have done small analysis on the various extended versions and submitted the report.
Then the major confusion from the business was, is it possible to upgrade without much effort and what would be the risk, is it the right path, do we have any alternaitive path etc.
We have drawn another beautiful diagram depicting all the available possibilities with risks and effort. Please find the same below.
The LEAST RISK solution is shown as the straight path, medium risk on top and HIGH RISK is shown as RED path.
We have shared a detailed list with Risks and Efforts on each path with ranking based on some quanitifiable values, which helped the business to arrive on a best solution.
I have drawn this on a paper, rewritten based on the discussions with the team and team was so happy to capture what I was talking about. I feel you also feel the same, and happy to hear your way of decision analysis.
http://en.wikipedia.org/wiki/Decision_analysis
http://home.ubalt.edu/ntsbarsh/opre640a/partix.htm
Let me share my recent experience how I helped myself to reach a solution and share the same to the business for taking easy decision.
Business requirement :
The "PRU web apps" which I would like to discuss is currently runs on IBM WAS 6.x platform with JSF 1.2. The business was forced to migrate the PRU web apps to IBM WAS 7.x platform as the IBM WAS 6.x EOL[End of Life] has reached nearby, which is Sept 13.
Problem :
Team had intiated their migration project and got halted in between as they got some migration issues due to the incompatibility of the JWL libraries[UI Widget libraries] of IBM JSF version of WAS 6.x and WAS7.x and the some of the code they have used in PRU web apps.
Initial Analysis :
Team has contacted IBM Support help, tried all options and finally left to live with the alternative options. Now the business got the red alert and started thinking multiple options, which includes -> Migrating to other JSF implementation, fixing the JWL compatibility issues, any hybrid options or even thinking little more futuristic and migrated to WAS 8.x with MyFaces2.0 which is the recent IBM platform and avoid another probable migration from 7.x etc etc.
Business has their own limitations : They cant use any other JSF implementations other than IBM specific and the PRU Web cant be deployed any other container other than IBM specific Application Servers.
What would be the best option or feasible option or shall we move little cautious way or phased migration.
Solution :
More than the solution, I would like to discuss on how we have arrived on a suitable solution or the best solution approach.
Since the major issue is the compatibility issue of IBM - JWL Widget library with update versions, we have intiated a prototype of replacing JWL libraries with Tomahawk libraries[MyFaces], which is a hybrid approach, and got it working reconfirming our initial analysis. Based on the prototype and long digging of various discussion forums/ IBM support, we have concluded to go for MyFaces 1.2 with WAS 7.x platform.
Now we have created a beautiful diagram of next solution approach. Please find the same below.
The Y-Axis says the future IBM WAS versions, and Y-Axis says the next MyFaces versions. The digram says, once we have reached the immediate solution, we should be able to move to any other combination without much effort.
We have done small analysis on the various extended versions and submitted the report.
Then the major confusion from the business was, is it possible to upgrade without much effort and what would be the risk, is it the right path, do we have any alternaitive path etc.
We have drawn another beautiful diagram depicting all the available possibilities with risks and effort. Please find the same below.
The LEAST RISK solution is shown as the straight path, medium risk on top and HIGH RISK is shown as RED path.
We have shared a detailed list with Risks and Efforts on each path with ranking based on some quanitifiable values, which helped the business to arrive on a best solution.
I have drawn this on a paper, rewritten based on the discussions with the team and team was so happy to capture what I was talking about. I feel you also feel the same, and happy to hear your way of decision analysis.
TERIMA KASIH ATAS KUNJUNGAN SAUDARA
Judul: Decision Analysis
Ditulis oleh Unknown
Rating Blog 5 dari 5
Semoga artikel ini bermanfaat bagi saudara. Jika ingin mengutip, baik itu sebagian atau keseluruhan dari isi artikel ini harap menyertakan link dofollow ke http://apk-zipalign.blogspot.com/2013/03/decision-analysis.html. Terima kasih sudah singgah membaca artikel ini.Ditulis oleh Unknown
Rating Blog 5 dari 5
Categories:
0 komentar:
Posting Komentar