I'm confused. Am I responding to "Data saving from FPGA" or "Save loop counter in an array or a text file"? And how did @majid12 respond to Ben before Ben posted?
I'm going to assume that the data are originally in FPGA "memory", and need to (eventually) be written in some fashion to a Disk file on a Host PC. No mention of the "monkey in the middle" (unless you have a PC with integral FPGA chips, something I know nothing about).
Do you know how to take data that you generate (say, using a random number generator and some While and For loops) and save it to disk? [If not, learn enough LabVIEW to be able to do this task, as you'll need these skills for this problem].
Do you know how to take data generated by code running on an FPGA and get it passed up to the system "hosting" the FPGA Chip? [In my experience, this has been what NI calls a "Real-Time Target" machine, running an Real-Time OS and having special "timing" hardware to give better "determinism" than is possible on a Windows multi-threaded PC]. Again, if you don't know how to do this, learn.
Finally, where Real-TimeTargets are ideal for Real-Time tasks (where "determinism" is important), PCs are often better for managing Data Acquisition and Data Processing, such as saving the data to hard drives or to the Web. One generally communicates between Host (PC) and Target (which has the FPGA) using TCP/IP -- one fairly robust method that NI provides is called Network Streams, which I use and can recommend (noting that I have not tried using TCP or UDP protocols directly).
Without knowing more about the details of what you want to do, it is difficult to make more specific recommendations. If you have code that you are trying to write and need help with it, by all means attach your code (please don't just send pictures of parts of a block diagram!) and you will get a response.
Bob Schor
