Re: Bit Packing (was: Re: HDF5 bitfields...)

Hi Ed,

> > > John Caron wrote:
> > > > the scale/offset can be calculated easily from the data itself. often, 
> > > > people want to apply different scale/offset to different parts of the 
> > > > same array, eg vertical levels.
> > > 
> > > and you replied:
> > > >     Hmm, how would you parameterize this?  Would a user select various 
> > > > parts
> > > > of the dataset's dataspace and specify scale/offset information for 
> > > > them?
> > > 
> > > When Harvey Davies was here from Australia for a visit about 8 years
> > > ago, we worked out two kinds of scaling for varying packing parameters
> > > along one or more dimensions of a variable: predefined scaling and
> > > adaptive scaling.  
> > > 
> > > With predefined scaling, the scale and offset values associated with a
> > > packed variable were stored in auxiliary arrays, varying along just
> > > the subset of dimensions used by these arrays.  For example, to store
> > > a packed array of temperatures, one might use
> > > 
> > >   dimensions:
> > >     time = ...
> > >     lat = ...
> > >     lon = ...
> > >     level = ...
> > >   variables:
> > >     byte temperature(time, level, lon, lat);
> > >     double temperature_scale_factor(level);
> > >     double temperature_add_offset(level);
> > > 
> > > which would use a possibly different (scale_factor, add_offset) pair
> > > for packing temperatures on each atmospheric level.  This would allow
> > > for greater precision using the same number of bits (or fewer bits for
> > > the same precision) than using one packing parameter pair for all the
> > > data, because this variable tends to have values that depend on level.
> > > It wouldn't work so well with other variables that don't have a
> > > level-dependence.
> > > 
> > > With adaptive scaling, the optimum scale and offset values were to be
> > > computed by the library for each slab of the variable as it was
> > > written, and stored in automatically-generated associated variables
> > > (or multidimensional attributes).
> > > 
> > > Although we defined interfaces for these types of scaling, they were
> > > never implemented.  Implementing adaptive scaling seemed pretty
> > > ambitious, and even the predefined scaling would have required
> > > adoption of new conventions for naming associated variables, etc.  And
> > > the proposals actually foundered on inability to agree on all the gory
> > > details, such as determining whether to permit the types of the
> > > scaling parameters to be user-specifiable in adaptive scaling, etc.
> >     Ok, I see.  Hmm...  I think that the adaptive scaling would actually be
> > somewhat easier that the predefined scaling you describe in HDF5.  With the
> > adaptive scaling, each chunk in the dataset could be scanned to compute the
> > optimum scale and offset values which would be stored with the chunk.  
> > Handling
> > predefined scaling that varied according to a position within the dataspace
> > seems like it would require accessing some information that was stored 
> > outside
> > each chunk and that might be a little unusual in the current implementation.
> > Predefined scaling that didn't vary across the dataspace would be easier 
> > than
> > either of those methods, of course.  Although it gets a little weird to 
> > define
> > any sort of scaling on non-numeric datatypes, we've got a mechanism for
> > disallowing that now.
> > 
> >     Quincey
> 
> Let me make sure I'm using the same terminology as the rest of you...
> 
> When we say "bit packing" we mean applying a scale and offset to a
> bunch of, for example 64-bit floats, and ending up with 32-bit floats?
> 16-bit ints as well?
> 
> When we say bitfields we mean storing data that represent actual
> bitfields, like the output of an 11-bit A/D converter in some
> scientific instrument. In that case we would not wish to apply
> scale/offset.
> 
> In the bitfield case we *would* like to store 11-bit values without
> having to round-up to 16-bits, but that is not what HDF5 bitfields
> provide. 
> 
> Have I got the terms correct? Because we have both a bitfield and a
> bit packing requirement:
> 
>  Bit packing
> 
> * Data may be bit packed.
    I agree with you, that it would be nice to allow packing the results of the
scale/offset operation, if possible.  You are making the same distinction that
I made earlier - that we are talking about two different methods for
"compressing" data, one by "packing out" unused bits (for bitfields) and another
by "scaling away" unused precision (and then possibly packing out any unused
bits).

    Quincey