Re: [ROOT] reduction

From: Victor Perevoztchikov (perev@bnl.gov)
Date: Thu Jan 17 2002 - 17:32:45 MET


Hi Christian,
> Hmm.  Does ROOT not compress Byte_t members?  Anyway, that is >50%
For each class stored Version = 2bytes, counter 4 bytes, and at leas one byte of data.
totally >= 7 bytes. 
>  use Byte_t instead of UShort_t
So you will save 1 byte which is 1/8 abot 12% not 50%

Victor

Christian Holm Christensen wrote:
> 
> Hi Rene,
> 
> On Thu, 17 Jan 2002 14:48:44 +0000
> Rene Brun <Rene.Brun@cern.ch> wrote
> concerning "Re: [ROOT] reduction":
> > Hi Christian,
> >
> > I must diasgree with you for at least 3 reasons.
> >   - ROOT compression will do better than your 1 byte data type.
> 
> Hmm.  Does ROOT not compress Byte_t members?  Anyway, that is >50%
> reduction; impressive.  However, in memory, you still loose 1 byte per
> object, which isn't a problem ofcourse :-)
> 
> >   - Very often , placing 1 Byte element in a class creates a misalignement
> >     problem and you run the risk to lose more space than you think to gain
> >     and also to lose time due to the misalignement
> 
> Whoops, forgot that issue.  But isn't that helped by the mechanism
> where you put all members of one type in one continues array (or
> something like that anyway).
> 
> >   - We cannot support all possible cases in the collection classes:
> >     a TObjArray with a wordcount of 1,2,4,8 bytes.
> 
> Well, I didn't intend the collection counter to be of various sizes.
> It was rather the objects that you put into the collection that could
> use Byte_t instead of UShort_t.
> 
> Yours,
> 
> Christian Holm Christensen -------------------------------------------
> Address: Sankt Hansgade 23, 1. th.           Phone:  (+45) 35 35 96 91
>          DK-2200 Copenhagen N                Cell:   (+45) 28 82 16 23
>          Denmark                             Office: (+45) 353  25 305
> Email:   cholm@nbi.dk                        Web:    www.nbi.dk/~cholm

-- 
Victor M. Perevoztchikov   perev@bnl.gov  perev@vxcern.cern.ch       
Brookhaven National Laboratory MS 510A PO Box 5000 Upton NY 11973-5000
tel office : 631-344-7894; fax 631-344-4206; home 631-345-2690



This archive was generated by hypermail 2b29 : Sat Jan 04 2003 - 23:50:39 MET