Re: Parallelism a la make -j <n> / GNU parallel

[Linda Walsh]

Ole Tange wrote:

> Can you explain how that idea would differ from sem (Part of GNU Parallel)?

----
        Because gnu parallel is written in perl?  And well, writing it in
perl.... that's near easy... did that about ... 8 years ago? in perl...
to encode albums in FLAC or LAME -- about 35-45 seconds/album...on my old 
machine.  But perl broke the script, multiple times .. (upgrades in perl)...

So am rewriting it...

        Doing it in shell... that would be a 'new' challenge... ;-)

        And people called me masochistic for trying to write complex
progs in shell ...

        Actually my first parallel encode was about 20 lines of shell...
but it just delayed launch of each job by some constant 'k', Wasn't too
efficient, as it usually ran too many jobs at once.

        I think I switched one using job control and keeping track of
the outstanding jobs using the jobs command.  With arrays and hashes (assoc 
arrays), it would be easier to be more flexible...

        maybe 'par' or something similar needs to be ported into a
"dll", that could be dropped into bash as an extension..?

        Hey... for that matter... um.. perl.dll... hmmmm...