Fragmentation question

(Sorry all, big brother encrypted my last response again=2E Retrying)=0D= =0A=0D=0A=0D=0A=0D=0AYou are fragmenting the index with this statement, but= you said it is=0D=0Athe primary key of your table=2E Can't you just fragm= ent the table this=0D=0Away? The index will get created implicitly and sto= red with the table,=0D=0Atherefore for this index new data will not cause t= wo separate actions,=0D=0Aone to store the data and one to store the index= =2E=0D=0A=0D=0AAnd it is always a good idea to have a remainder clause even= if you=0D=0Acan't foresee it ever being used=2E=0D=0A=0D=0ADid you mention= there is more than one index on this table? Any index=0D=0Athat is not pa= rt of the fragmentation strategy will take a very long=0D=0Atime to add=2E = You should consider if you can remove this second index,=0D=0Abut if not y= ou definitely need to disable that index during data inputs=0D=0Aand rebuil= d after the data is completed=2E=0D=0A=0D=0AThanks,=0D=0AKate Tomchik, IT A= rchitect=0D=0A=0D=0A-----Original Message-----=0D=0AFrom: informix-list-bou= nces@iiug=2Eorg=0D=0A[mailto:informix-list-bounces@iiug=2Eorg] On Behalf Of= Jarrod Teale=0D=0ASent: Monday, December 15, 2008 9:58 PM=0D=0ATo: informi= x-list@iiug=2Eorg=0D=0ASubject: Attach fragment to index not atomic=0D=0A= =0D=0AHi,=0D=0AIDS11=2E10=2EUC2W2 on RHEL4 =0D=0A =0D=0AWe have the index (= and a few more like it) below=2E It is used for the=0D=0Aprimary key of a t= able=2E=0D=0AWhen I attach a new fragment to the system it reads every page= of the=0D=0Aindex=2E So adding 12 fragments, one for each month is a pain= =2E=0D=0A alter fragment on index pk_tagdata_73 add (dt_filter =3D 20090= 1 ) in=0D=0Astan_03a_idx_2009_01 before stan_03a_idx_2008_01=0D=0A =0D=0AAs= you can see from the schema, there is no remainder clause, so I would=0D= =0Aexpect this operation to be atomic=2E The fragments for the table are=0D= =0A(same fragment scheme)=2E=0D=0A =0D=0AThere is no data in the table for = the fragments being added (the date=0D=0Afilter is for Jan 2009)=0D=0AAny i= deas why this is not running as an atomic operation? It's taking a=0D=0Area= lly long time over 5 TB data!=0D=0A =0D=0Acreate unique index "sdev"=2Epk_t= agdata_73 on "sdev"=2Etagdata_73=0D=0A (group_id,dt_filter,sample_dt) us= ing btree=0D=0A fragment by expression=0D=0A (dt_filter =3D 200801 ) in= stan_03a_idx_2008_01 ,=0D=0A (dt_filter =3D 200802 ) in stan_03a_idx_20= 08_02 ,=0D=0A (dt_filter =3D 200803 ) in stan_03a_idx_2008_03 ,=0D=0A = (dt_filter =3D 200804 ) in stan_03a_idx_2008_04 ,=0D=0A (dt_filter =3D = 200805 ) in stan_03a_idx_2008_05 ,=0D=0A (dt_filter =3D 200806 ) in stan= _03a_idx_2008_06 ,=0D=0A (dt_filter =3D 200807 ) in stan_03a_idx_2008_07= ,=0D=0A (dt_filter =3D 200808 ) in stan_03a_idx_2008_08 ,=0D=0A (dt_= filter =3D 200809 ) in stan_03a_idx_2008_09 ,=0D=0A (dt_filter =3D 20081= 0 ) in stan_03a_idx_2008_10 ,=0D=0A (dt_filter =3D 200811 ) in stan_03a_= idx_2008_11 ,=0D=0A (dt_filter =3D 200812 ) in stan_03a_idx_2008_12 ,=0D= =0A (dt_filter =3D 200701 ) in stan_03a_idx_2007_01 ,=0D=0A (dt_filte= r =3D 200702 ) in stan_03a_idx_2007_02 ,=0D=0A (dt_filter =3D 200703 ) i= n stan_03a_idx_2007_03 ,=0D=0A (dt_filter =3D 200704 ) in stan_03a_idx_2= 007_04 ,=0D=0A (dt_filter =3D 200705 ) in stan_03a_idx_2007_05 ,=0D=0A = (dt_filter =3D 200706 ) in stan_03a_idx_2007_06 ,=0D=0A (dt_filter =3D= 200707 ) in stan_03a_idx_2007_07 ,=0D=0A (dt_filter =3D 200708 ) in sta= n_03a_idx_2007_08 ,=0D=0A (dt_filter =3D 200709 ) in stan_03a_idx_2007_0= 9 ,=0D=0A (dt_filter =3D 200710 ) in stan_03a_idx_2007_10 ,=0D=0A (dt= _filter =3D 200711 ) in stan_03a_idx_2007_11 ,=0D=0A (dt_filter =3D 2007= 12 ) in stan_03a_idx_2007_12 ;=0D=0A=0D=0A =0D=0AThanks in advance for the = help=0D=0A =0D=0A=0D=0AJarrod Teale=0D=0A=0D=0ATeam Lead - Manufacturing Ex= ecution Systems=0D=0A=0D=0AAutomation & Process Control Group=0D=0A=0D=0A= =0D=0A-----------------------------------------=0D=0AThe information contai= ned in this e-mail and any attached documents=0D=0Amay contain information = that is confidential or otherwise protected=0D=0Afrom disclosure=2E If you = are not the intended recipient of this=0D=0Amessage, or if this message has= been sent to you in error, please=0D=0Aimmediately alert the sender by rep= ly e-mail and then delete this=0D=0Amessage, including any attachments=2E A= ny dissemination, distribution=0D=0Aor other use of the contents of this me= ssage by anyone other than=0D=0Athe intended recipient is strictly prohibit= ed=2E=0D=0A