Mainframe vs .NET/J2EE
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
Personally, I think J2EE is highly complex. Make sure you have done your studies before you dive into it. I'd say it requires a project or two before you get the hang of best practices, tools and environments. It's a jungle. J2EE is not the oh-so-easy silver bullet that Sun would like you to believe. I'm not saying that .NET is any less complex. If anything, it's probably more a moving target than what J2EE is. There's a new TLA coming out of Redmond every month, obsoleting the TLAs from the month before. On the other hand, Microsoft is probably the best company I know of that keeps backward compatibility.
-- Kein Mitleid Für Die Mehrheit
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
Jwalant Natvarlal Soneji wrote:
1. If all mainframe applications should be rewritten in .NET/J2EE?
No. There's generally very few compelling reasons that an application that does the job it's supposed to should be rewritten. Plus, why only .NET/J2EE? These aren't the only choices available, and it says a lot about the "industry" that these are seen as the only choice - there are many areas where these are not viable solutions.
Jwalant Natvarlal Soneji wrote:
2. What are the benefits a business unit can get?
Again, this is a tough one. The obvious "benefit" is staffing - younger coders tend not to have skills in the languages that are typically used on mainframes, but this can be mitigated by having in-house training and proper mentoring.
Jwalant Natvarlal Soneji wrote:
3. Is it feasbile?
No. Again, why fix something that isn't broken.
Jwalant Natvarlal Soneji wrote:
4. Any alternatives?
You can probably guess what my answer is going to be here. Leave it alone. BTW - this sounds suspiciously like homework.
"WPF has many lovers. It's a veritable porn star!" - Josh Smith
As Braveheart once said, "You can take our freedom but you'll never take our Hobnobs!" - Martin Hughes.
-
Jwalant Natvarlal Soneji wrote:
1. If all mainframe applications should be rewritten in .NET/J2EE?
No. There's generally very few compelling reasons that an application that does the job it's supposed to should be rewritten. Plus, why only .NET/J2EE? These aren't the only choices available, and it says a lot about the "industry" that these are seen as the only choice - there are many areas where these are not viable solutions.
Jwalant Natvarlal Soneji wrote:
2. What are the benefits a business unit can get?
Again, this is a tough one. The obvious "benefit" is staffing - younger coders tend not to have skills in the languages that are typically used on mainframes, but this can be mitigated by having in-house training and proper mentoring.
Jwalant Natvarlal Soneji wrote:
3. Is it feasbile?
No. Again, why fix something that isn't broken.
Jwalant Natvarlal Soneji wrote:
4. Any alternatives?
You can probably guess what my answer is going to be here. Leave it alone. BTW - this sounds suspiciously like homework.
"WPF has many lovers. It's a veritable porn star!" - Josh Smith
As Braveheart once said, "You can take our freedom but you'll never take our Hobnobs!" - Martin Hughes.
Thanks Pete,
Pete O'Hanlon wrote:
this sounds suspiciously like homework.
It is not actully. :sigh: I am trying to understand if there is any big differences could be achived if mainframe is migrated to these languages.
Jwalant Natvarlal Soneji, BE IT, India
-
Thanks Pete,
Pete O'Hanlon wrote:
this sounds suspiciously like homework.
It is not actully. :sigh: I am trying to understand if there is any big differences could be achived if mainframe is migrated to these languages.
Jwalant Natvarlal Soneji, BE IT, India
Jwalant Natvarlal Soneji wrote:
I am trying to understand if there is any big differences could be achived if mainframe is migrated to these languages.
Yes - you'd cost a company millions for no real benefit.
"WPF has many lovers. It's a veritable porn star!" - Josh Smith
As Braveheart once said, "You can take our freedom but you'll never take our Hobnobs!" - Martin Hughes.
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
Rewrites for the sake of rewrites are stupid. Unless there're major changes needed; or the app's decayed to the point of being unmaintainable a rewrite would be a poor use of resources.
The European Way of War: Blow your own continent up. The American Way of War: Go over and help them.
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
Jwalant Natvarlal Soneji wrote:
Its about rewriting mainframe apps in .NET/J2EE.
There are new applications emerging which replace existing mainframe applications. For example, the software I am working on does the same thing. The reasons are following:- 1. Business requirements are changing 2. Existing mainframe systems are not flexible to accommodate the changes
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
I'm in such an environment right now so I can relate somewhat. The old mainframe system does what it is supposed to, so there is no immediate benefit of porting anything. The real benefit would come from new possibilities, e.g. the systems where I am were designed some 40 years ago and have been patched and patched over and over and have now become hard to change and maintain. Design decisions from decades ago make it impossible to add new features w/o massive rewrites. (and lack of testability will most likely cause new bugs to surface in the production environment) So a .NET or Java rewrite could make it much easier to add new features and maintain. Due to better tools, testability and easier to model the problem domain in OOP. //Roger
-
All the big financial institutions would crash. If it ain't broke, don't break it.
PIEBALDconsult wrote:
If it ain't broke, don't break it.
Also if it's broken, don't break it. :laugh:
Wout
-
Jwalant Natvarlal Soneji wrote:
I am trying to understand if there is any big differences could be achived if mainframe is migrated to these languages.
Yes - you'd cost a company millions for no real benefit.
"WPF has many lovers. It's a veritable porn star!" - Josh Smith
As Braveheart once said, "You can take our freedom but you'll never take our Hobnobs!" - Martin Hughes.
Pete O'Hanlon wrote:
Yes - you'd cost a company millions for no real benefit.
Agreed. The benefit is not in changing the code, but in changing the hardware. If you are not leveraging the high transaction capability of the Mainframe, you ought to ditch it. Replace it with a scalable server farm. Something I have personally done at several facilities, such as Hospitals. The annual support overhead and cost of keeping Mainframe Sys Admins around is huge for mainframes (along with the required Annual Support cost from the manufacturer). You need to justify that cost first. It certainly can be justified, but in a lot of cases it is not.
-
Its about rewriting mainframe apps in .NET/J2EE. What do you think? 1. If all mainframe applications should be rewritten in .NET/J2EE? 2. What are the benefits a business unit can get? 3. Is it feasbile? 4. Any alternatives?
Jwalant Natvarlal Soneji, BE IT, India
Many, perhaps most, mainframes apps are high availability. IBM already has a program to move these to AIX and RS6000 systems.