Can I write to a buffer in parallel using some set of compute pipelines?

[dalduba]

The issue is to build some big buffer from a sequence of other buffers (N>limit of max storage buffers per shader). The building process is not trivial, it is more complex than just concatenation. I have a shader that puts the data from some input buffer to concrete place in out buffer, that depends on the current input buffer. Thus there is no collisions in out data ranges if I run N pipelines for N input buffers.
Here was an answer about pipelines parallel behavior. #2999

And I suggest that there is no chance to call the pipelines in parallel.

Is this right? Are there any suggestions?