[GRLUG] FYI - old mysql and new gear

[Bob Kline]

OK, the answer is of course hardware.
The illusion, as it's called, of running
more than one program at a time started
with the time slice, which is as old as
Unix itself.  And even that might not have
been cut from whole cloth.  Context switching
is necessary in order to suspend a process,
which your piece says is modernese for
program. That function is as old as
interrupts, and there is perhaps a connection
between interrupts and time slicing, where
the "interrupt" is a signal saying a time
slice has run out of time.

Anyway,  it gets back to the notion of
keeping as many balls in the air at the
same time.  Modern processors can
run more than one thread at the same
time, but that might be a way of saying
one has more than one processor in
the same core.  One still cannot get
around the basic limitations of branches,
and the other kinds of hardware management
that a CPU has always had to do.  The
CDC 6600 had a dozen or so satellite
processors for doing arithmetic processing.
The problem was still management.

The lingo changes, but apparently not
the problems very much.  Let me ask
how much threading speeds things up
in practice?  Sun Microsystems used
"lightweight threads," or "lightweight
processes" 20 or more years ago. More
fine grained control.  I suspect the
concepts have made their way in to
some hardware today.  Something like
the 35 year old notion of a math
coprocessor - e.g., the old 80287.
Eventually that all made its way on to
the CPU, but it's more integration than
evolution.

IMHO anyway.

   -- Bob


On Thu, Sep 22, 2011 at 3:22 PM, Michael Mol <mikemol at gmail.com> wrote:

> On Thu, Sep 22, 2011 at 3:16 PM, Bob Kline <bob.kline at gmail.com> wrote:
> > Is hyperthreading more of a hardware
> > function or a software?  The thing that
> > often kills speedup is branching.  If you
> > can keep a long sequence of computations
> > going, you get speedup, but that doesn't
> > always work out in practice either.
> > What role do compilers have here, if any?
> >    -- Bob
>
>
> http://web.archive.org/web/20030811130623/http://arstechnica.com/paedia/h/hyperthreading/hyperthreading-1.html
> --
> :wq
>
> --
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
>
> _______________________________________________
> grlug mailing list
> grlug at grlug.org
> http://shinobu.grlug.org/cgi-bin/mailman/listinfo/grlug
>

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://shinobu.grlug.org/pipermail/grlug/attachments/20110922/e5a843ee/attachment-0001.html>