Microsoft Makes Push for COBOL Migration
geoff313 writes: "It would appear that Microsoft
is making a real push for the migration of existing COBOL applications to Windows and their .Net platform. Micro Focus, a company who makes
COBOL migration products and last year became a member of Microsoft's Visual
Studio Industry Partner (VSIP) program, announced their Net
Express with .Net product, a plug-in to Microsoft Visual Studio .Net
2003. It allows for COBOL code to be integrated and manged with other code in Visual Studio. In an interview with eWeek he declares that 'Micro Focus and Microsoft are bringing the mainframe to Windows and .Net'. This makes me wonder, are there any Open Source projects working to provide for this eventual migration? Gartner estimates that over 75% of business data is processed by an approximately 200 billions lines of COBOL, so
this seems like a huge potential market to lose to Microsoft."
What you have to understand is that the reputed stability and scalability of mainframe systems is largely a cultural phenomenon within business. Though the hardware itself is rock-solid, and the software has been debugged over a period of 40+ years in some cases, it still is quite primitive underneath. For anyone who looks at a mainframe (i.e. System 370 MVS) architecture after using a 64-bit Unix, or even NT, its really quite scary. For example, you have to tell the OS how big your file is ever going to get before you can write to it. Files are usually record oriented with fixed maximums like 80 characters per line (this comes from punched cards). File allocations and volumes and are constantly on the verge of filling up without admin intervention. All names are limited to 8 uppercase characters. And so on.
Of course there are additional utilities that can handle much of this automatically, but configuring them means learning an even larger set of complex utilities with even more options. And it still requires knowledge of the underlying implementation model.
So when a business decides it's big enough to need a mainframe, it fully expects to have a full-time staff dedicated to servicing it. Operators to handle print, tape, and job scheduling. Regular hardware maintenance from the manufacturer. Systems analysts and programmers to configure and run the utilities. And they're also educated not to expect any significant processing to happen faster than overnight! "Interactive" sessions can have a response time of 2 or 3 seconds.
On the other hand, if you propose a system based on PC-compatible systems (no matter how powerful), they expect to be able to hire the staff straight out of the local community college. They expect a ratio of servers to admins much greater than one.
PC and Unix hardware has also gotten darned reliable, even if the mainframe has an extra 9 or two on it.
IBM in its newer systems is moving more and more functionality into the unix side (it comes with a unix runtime layer built-in), while also marketing running multiple instances of Linux simultaneously with the traditional OSes.