Re: [Ganglia-developers] Bugzilla Bug 193: Avg Load percentages and overall cluster utilization.
On Aug 21, 2008, at 9:49 AM, Bernard Li wrote: Hi Martin: On Wed, Aug 20, 2008 at 12:35 AM, Martin Knoblauch [EMAIL PROTECTED] wrote: thanks for spotting. It just shows that the number of deployment scenarios is just to big for the patch/feature developers. And that we cannot assume that a release will have been completely tested for all scenarios. Yeah, we really need to motivate more users to help with testing -- maybe we can give out free t-shirts for finding bugs during the testing cycle :P Alternatively, we can rollback this backport and punt it until 3.1.2. No real strong feelings. I just rolled back c1694. I'll work on fixing the code for 3.1.2. On a related note, I think we should distinguish between a Grid and a Meta-Grid (i.e. a grid of grids) in the Front End -- do people care? Definitely a good idea, as it seems to be a more and more common case. So are we all okay with the term Meta-Grid or can we think of a fancier name, maybe like Cloud or Constellation or Galaxy, or something like that ;-) I prefer Grid of Grids (or gog); then when we have a massive grid of (grids of grids), we can have Magog to go with it (http://en.wikipedia.org/wiki/Gog_and_Magog ), at which time it's appropriate to drink a bunch of stout call it a day :-) More seriously: at some point in the hierarchy, you have to just say container and let containers contain containers or you can't build arbitrary numbers of levels. If I understand current Ganglia semantics: Clusters contain hosts Grids contain clusters Grids of grids (metagrids) contain grids What contains metagrids? I'd be OK with the word metagrids IFF the answer to that is metagrids. Turtles, turtles, turtles -- -- ReC Cheers, Bernard - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
Re: [Ganglia-developers] Bugzilla Bug 193: Avg Load percentages and overall cluster utilization.
- Original Message From: Bernard Li [EMAIL PROTECTED] To: ganglia-developers ganglia-developers@lists.sourceforge.net Sent: Wednesday, August 20, 2008 2:33:02 AM Subject: [Ganglia-developers] Bugzilla Bug 193: Avg Load percentages and overall cluster utilization. Dear all: Hi Bernard, http://bugzilla.ganglia.info/cgi-bin/bugzilla/show_bug.cgi?id=193 This patch has not been tested with meta-sources (i.e. gmetad aggregating gmetad) and thus the average utilization numbers are incorrect for grid of grids. thanks for spotting. It just shows that the number of deployment scenarios is just to big for the patch/feature developers. And that we cannot assume that a release will have been completely tested for all scenarios. I currently have an incomplete fix, but I need to get consensus as to what average utilization really means for grid of grids: should average utilization for a grid be load average divided by the number of cpus for the *entire* meta-grid or just over the grid in question? Alternatively, we can rollback this backport and punt it until 3.1.2. No real strong feelings. On a related note, I think we should distinguish between a Grid and a Meta-Grid (i.e. a grid of grids) in the Front End -- do people care? Definitely a good idea, as it seems to be a more and more common case. Cheers Martin - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
Re: [Ganglia-developers] Bugzilla Bug 193: Avg Load percentages and overall cluster utilization.
On Tue, Aug 19, 2008 at 20:33, Bernard Li [EMAIL PROTECTED] wrote: I currently have an incomplete fix, but I need to get consensus as to what average utilization really means for grid of grids: should average utilization for a grid be load average divided by the number of cpus for the *entire* meta-grid or just over the grid in question? For a both the grid and meta-grid I think the average utilization should be counted for all hosts. Consider a meta-grid that gets data from from two grids A and B. Grid A has 100 hosts and Grid B has 20. If 15 machines are running full-tilt (e.g. 100% utilization on 15 different hosts), then the Grid utilization figures are quite different. For Grid A, there's a utilization of 15%. For Grid B, it's 75%. If we weight the two grids equally, we get an average utilization of 82.5%, even though there are 90 idle systems. On the other hand, if you take into account the number of hosts, you get a different figure: (15+15)/(100+20) = 25% average utilization. To me, this seems to be a more accurate representation of the overall usage. Now, this does *not* take into account things like relative CPU speeds, or multi-core systems. 100% utilization on my ancient Celeron is quite a bit different than 100% on the latest quad-core Operton. I don't think that we need to delve quite that far into compariing Alternatively, we can rollback this backport and punt it until 3.1.2. Probably the cleanest solution for now. On a related note, I think we should distinguish between a Grid and a Meta-Grid (i.e. a grid of grids) in the Front End -- do people care? Yes, it would be good to distinguish between them, even if all it does is say Meta-Grid instead of Grid. -- Jesse Becker GPG Fingerprint -- BD00 7AA4 4483 AFCC 82D0 2720 0083 0931 9A2B 06A2 - This SF.Net email is sponsored by the Moblin Your Move Developer's challenge Build the coolest Linux based applications with Moblin SDK win great prizes Grand prize is a trip for two to an Open Source event anywhere in the world http://moblin-contest.org/redirect.php?banner_id=100url=/ ___ Ganglia-developers mailing list Ganglia-developers@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/ganglia-developers
