On Tue, 9 Jan 2018, Jan Fajerski wrote: > On Tue, Jan 02, 2018 at 04:54:55PM +0000, John Spray wrote: > > On Tue, Jan 2, 2018 at 10:43 AM, Jan Fajerski <jfajerski@xxxxxxxx> wrote: > > > Hi lists, > > > Currently the ceph status output formats all numbers with binary unit > > > prefixes, i.e. 1MB equals 1048576 bytes and an object count of 1M equals > > > 1048576 objects. I received a bug report from a user that printing object > > > counts with a base 2 multiplier is confusing (I agree) so I opened a bug > > > and > > > https://github.com/ceph/ceph/pull/19117. > > > In the PR discussion a couple of questions arose that I'd like to get some > > > opinions on: > > > > > - Should we print binary unit prefixes (MiB, GiB, ...) since that would be > > > technically correct? > > > > I'm not a fan of the technically correct base 2 units -- they're still > > relatively rarely used, and I've spent most of my life using kB to > > mean 1024, not 1000. > We could start changing the "rarely used" part ;) But I can certainly live > with keeping the old units. > > > > > - Should counters (like object counts) be formatted with a base 10 > > > multiplier or a multiplier woth base 2? > > > > I prefer base 2 for any dimensionless quantities (or rates thereof) in > > computing. Metres and kilograms go in base 10, bytes go in base 2. > > > > It's all very subjective and a matter of opinion of course, and my > > feelings aren't particularly strong :-) > As far as I understand the standards regarding this (IEC 60027, ISO/IEC 80000, > probably more) are talking about base 2 units for digital data related units > only. I might of course misunderstand. > What is problematic I find is that other tools will (mostly?) use base 10 > units for everything not data related. Say I plot the object count of ceph in > Grafana. It'll use base 10 multipliers for a dimensionless number. Since > Grafana (and I imagine other toolsllike this) consume raw numbers we'll end up > with Grafana displaying a different object count then "ceph -s". Say 1.04M vs > 1M. Now this is not terrible but it'll get worse with higher counts quickly. > In the original tracker issue it's noted that this was reported with cluster > containing 7150896726 objects. The difference from grafana to "ceph -s" was > 7150M vs 6835M. Right. I find the *iB units annoying myself, and I'm not sure I'll ever be able to say "pebibyte" out loud, but I can't think of a good reason not to be correct and precise. As a practical matter, I wonder if the PR should eliminate si_t entirely and replace it with dec_si_t and bin_si_t. Or, since the binary units aren't actually SI units, replace si_t with dec_si_t (to be explicit!) and bin_unit_t, or {dec,bin}_unit_t, or similar. I suspect that si_t vs iec_t or similar won't be sufficient for the developer to choose the right thing. sage _______________________________________________ ceph-users mailing list ceph-users@xxxxxxxxxxxxxx http://lists.ceph.com/listinfo.cgi/ceph-users-ceph.com
 |