[rescue] Sun 711

[James Lockwood]

On Thu, 2 May 2002, David Passmore wrote:

> Take a look at how fast the bus and memory bandwidths are on the latest
> Pentium IV motherboards with RDRAM. Sure, it's the same old bus, but the
> clocks have been ratcheted up to insane (well, relative to what they were a
> couple of years ago) levels. You don't need a crossbar switch for this
> simple task.

Look at STREAMs numbers.  300MB/s both coming in and going out will sap
half of your RAM bandwidth, minimum.  This is not insignificant.  The
clocks are not appreciably higher in new PC RAM technologies.
Synchronously clocked data transfers per clock have gone up, though.

Other thoughts: is this going through a windowing system?  Even if it can
be blitted to the screen "DGA-style", does it have to be done line by
line?  This kind of start/stop behaviour is poison to modern memory
systems, they heavily rely on main RAM being predominantly used for
sequential fills.

I've built servers to handle this kind of throughput and I stand by my
statement.

> Interrupts will kill you? Interrupts from what? We're not talking about a
> multi-purpose machine here. There's a single task running, the machine is
> probably doing a lot of DMA transfers across the bus, and that's about it.

You're on x86, no DVMA, and the framebuffer is probably on a different
channel (AGP).  CPU has to intervene to blast data out to the framebuffer
every 1/24 second (realistically you need more margin), so you can't use
huge DMA transfers locking everything up.  You pay an interrupt overhead
per transfer, and it becomes highly significant once you have a few
hundred MB/s flowing.  Big database servers smack into this all the time,
again running only one or two tasks.  Saturate a 64/66 PCI bus sometime
and watch your interrupt usage.

This is where mainframe channel controllers capable of offloading entire
lists of I/O operations shine.

-James