Paul,

Why would you consider gmetrics to be "second class"? I find those to be
much more useful than the builtin metrics. In fact, the only thing
that's nice about builtin metrics (other than the fact that you get them
out-of-the-box) is that they get reported even when the machine is under
extremely high load.


I view Ganglia as a framework rather than a performance monitoring
solution. Anything can be encoded in the UDP messages through the use of
gmetric, and specialized web interfaces can be quite easily built
through the use of the "custom metrics addon" I'd written a while back (
http://wtf.ath.cx/screenshots.html ). So you can not only access the XML
data on machine, cluster & grid levels, you can also generate whatever
UIs you want for your users... regardless of whether the metrics are
hardcoded or brought in from gmetric.


The only thing left on my wishlist is support for 64bit metrics, to
achieve better scalability over aggregated data.


Cheers,

Alex



Paul Millar wrote:

> Hi Richard,
>
> On Wednesday 07 February 2007 16:11, [EMAIL PROTECTED] wrote:
>   
>> a swiss consultancy has implemented a native windows gmond
>>     
> [...]
>
> Thanks for the announcement.
>
> Most HEP people run Linux clusters, but I know people that like Ganglia and 
> run Windows machines; I'll forward the link.
>
> [...]
>   
>> The final point is that as the extra metrics are binary coded, it should
>> be deployed in an all or nothing way per cluster.
>>     
>
> This is a shame.  So one cannot simply run the nice gmond.exe on a WIndows 
> machine and start monitoring it within an existing *nix-based ganglia 
> framework.
>
> If I've understood this correctly, this reflects (what I feel is) one of the 
> weaknesses of Ganglia at the moment: non-core data (e.g. gmetric) is treated 
> as if it were second-class information.  If other gmond-like daemons are to 
> send other interesting metrics, then ganglia should move away from 
> hard-coding the core metrics into the binary encoding (protocol.x).
>
> Instead, ganglia's encoding of data could be metric neutral;  a UDP packet 
> would contain multiple metrics,  all, some or none of which could be "core".  
> What are currently core metrics, as provided by gmond, could be identified by 
> their names (i.e. have "well known" names).
>
> Is this in keeping with any current plans?
>
> Cheers,
>
> Paul.
>
>   
> ------------------------------------------------------------------------
>
> -------------------------------------------------------------------------
> Using Tomcat but need to do more? Need to support web services, security?
> Get stuff done quickly with pre-integrated technology to make your job easier.
> Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
> http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
> ------------------------------------------------------------------------
>
> _______________________________________________
> Ganglia-developers mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/ganglia-developers
>   

Reply via email to