No worries ... it's a good and needed dialogue, and (in my opinion)
points out the need for flexibility in reporting in ganglia.
I'm just trying to convince my boss we really don't HAVE that many
CPUs! Seeing is believing, I guess. :->
rob
On Feb 16, 2005, at 9:35 AM, Sean Dilda wrote:
Robert E. Parrott wrote:
Maybe I'm missing something here but full load is NOT 4 processes
with hyperthreading.
Hyperthreading is a mixed bag, and while great for Desktop use, for
compute & memory bandwidth intensive jobs, can be a real drain,
because both processes are computing for the same cache. Most science
apps most tend to be memory bandwidth intensive, and gain from larger
cache. Thus, despite the grandiose claims of Intel, hyperthreading
is a problem here.
I agree completely. Which is why I assumed that by leaving HT on you
decided it was the best option for your cluster and actually wanted
jobs to use the virtual processors.
In fact, the actual throughput (i.e. in condor sense) of total Mops
with & without hyperthreading is often lower with hyperthreading on
... the need to schedule to the processes on the same core often
causes bottlenecks that slow down the processing. Thus we often have
limited our system to running 1 process per CPU. However, there are
times/applications where we would benefit from having hyperthreading
enabled, and need to have that as an option, without rebooting the
whole cluster.
I.e. "full load" in the load average reporting sense is in reality
dependent on the application, not on whether the OS reports 1 or 2
processors. Thus it would be quite useful to make this configurable
in just a small way.
Now I see. That is a tricky setup. In my case, we disabled HT in the
entire cluster due to the reasons stated above, as well as the fact
that there's only one fpu per physical processor.
So .... I take it that there is no such option (my original
question), since the entire discussion has been about whether I
really should want this or not, and not whether it is possible! :->
Unfortunately, I don't know of an option that'll do that.
I'm sorry if my question bothered you. I was just trying to figure
out why you thought ganglia was doing the wrong thing, when I thought
it was doing the right thing. I think I understand now.
