Electrical – Hard drive place in Processor Memory map

This sounds like a job for embedded Linux.

1. Persistent file system. Forget IDE (save yourself 40 pins) and go for a board that uses a flash card.
2. More RAM and Flash. Typical embedded Linux boards have RAM in the megabytes.

As for peripherals, driving 40 servos could be a question here on its own. How are you doing this now? For the rest of your peripheral requirements, here's a board that seems to fit that has a good community as well:

http://beagleboard.org/static/beaglebone/latest/Docs/Hardware/BONE_SRM.pdf

The tool chain has a C++ compiler, it has SPI, UARTs, and even a PWM. This is what's being claimed in the PDF at least, you'll have to make sure that there are drivers for all those peripherals available to you at the application level for whichever distro of Linux that you put on. Hopefully the one they provide has everything you need.

So basically, if you can port whatever you've written to a Linux PC, there's a good chance you can port it to an embedded linux target. However, I'm willing to bet that if all you're using from C++ is <fstream> and <string>, you could probably do a C re-write and save yourself the overhead of Linux.

It's not the width of number it can store, it's the width it can work with in a single operation. Customarily (but not necessarily) this also has a degree of correlation to the width of native memory addressing, and thus the amount of storage which can be easily mapped without ugly workarounds such as segmentation or bank switching.

Today's 32-bit cores are superior to 8-designs in most respects (flexibility, flat memory model, and of course performance), with the major exceptions being legacy systems, applications with extreme volume and price pressure (otherwise pricing tends to correlate better with on chip memory size than with core width), and side effects of process/density. The later can provide things like 5v operation, or possibly in some cases greater radiation hardness or a simplicity advantage if trying to prove the CPU design itself to be free of logic errors. One final process/age side effect of value to many hobbyists is that 8-bit cores in DIP packages are common, while 32-bit devices in such packages are rarer (though they do exist).