Re: Queries

[Alex Sim <ASim@lbl.gov>]

I just ran a preliminary test and got the following result from QE.
It's little different from David Z.'s estimate...

Q1:
mean_eta<-0.2 processing time : 30
QE returned fileidset.length = 62 .maximum=123
David Z estimated   100 events in 55 files.
100/55 =~ 2 * 30 (pt) = 60 (seconds processing time per file)
55 * 150 =~ 8250 seconds to finish the query

Q2:
glb_trk_tot>5000&mean_pt>0.5&tagrand>1000  processing time :20
QE returned fileidset.length = 58 .maximum=109
David Z estimated   108 events in 80 files
108/80 = 1.35 * 20 = 27 (seconds processing time per file)
80*150 = 12000 seconds to finish the query

Q3:
glb_trk_tot>0&glb_trk_tot<10&n_vert_total<3  processing time: 50
QE returned fileidset.length = 40 .maximum=73
David Z estimated   40 events in 29 files
40/29 =~1.38 * 50 = 69 (seconds processing time per file)
29*150 = 4350 seconds to finish the query

Q4:
glb_trk_tot>0&glb_trk_tot<10&n_vert_total<3  processing time: 60
QE returned fileidset.length = 40 .maximum=73
David Z estimated  73 events in 40 files
73/40 =~ 1.825 * 60 = 109.5 (seconds processing time per file)
40*150 = 6000 seconds to finish the query

Q5:
glb_trk_tot>0&glb_trk_tot<10&mean_eta>0.2  processing time: 90
QE returned fileidset.length = 20 .maximum=24
David Z estimated 24 events in 20 files
24/20 = 1.2 * 90 = 108 (seconds processing time per file)
20 * 150 = 3000 seconds to finish the query

-- 
Alex Sim                        
MS 50B-3238                      http://sim.lbl.gov
DM, NERSC, LBNL                        ASim@lbl.gov
1 Cyclotron Road             office: (510) 495-2290
Berkeley, CA 94720              FAX: (510) 486-4004