Re: [Lse-tech] topology design notes

[Paul J. <pj...@sg...>]

On Mon, 26 Nov 2001, Niels Christiansen wrote:
> How can we take your ideas and expand them to cover all kinds of
> Linux exploration including topology and instrumentation of
> things like kernel- and device driver statistics, locks, everything?

So we have need of topology for (at least):
    NUMA-aware tools and system services
    NUMA-aware scheduler and memory allocation refinements
    System management and performance instrumentation
    Plug-and-Play and hot swap
    Field service and diagnostics
    Robust configuration in changing configurations (SGI's ioconfig)
    Power Management

In the particular case of "System management and performance
instrumentation", I suspect that SGI will be more focused on
its own solution here - Performance Co-Pilot:

    http://www.sgi.com/software/co-pilot/

In the case of hot swap, it seems that some folks from NEC
are closer to worrying about solutions, just based on what
I see passing by on public lists such as this.

However, one of your points is an excellent one.  Topology is not
just for NUMA.

As for using the same framework (kernel hooks, system calls,
daemon for presentation) for instrumentation such as driver
statistics and lock metering, there I am more cautious, and
tend to one of two extremes:
  1) each such instrumentation hook has its own apparatus, or
  2) a common high level framework of a fairly powerful nature.

I see lock metering as an example of (1) - see:
    http://oss.sgi.com/projects/lockmeter/
and Performance Co-Pilot (above) as an example of (2).

I doubt that the simple framework I suggested as a radical
alternative for organizing topology discovery would be appropriate
for all these other uses.

But I do think that all these other uses do have need for
exploring a systems topology.


                          I won't rest till it's the best ...
                          Manager, Linux Scalability
                          Paul Jackson <pj...@sg...> 1.650.933.1373