I have a small problem. I would like to build a database containing a
main Tree with, say, 30 Million events, labelled by their start time.
These events arrive at random times, and their start times are not
sequential, i.e. do not correspond to arrival time.
There should be also another tree featuring an index with increasing
time (pair index-in-the-main-tree + start-time), so I can access quickly
the information.
Now, this index needs to be sorted. Since it has at least 30 M entries
(I would like 100 M), it cannot be sorted in memory (around 1/2 Gig). Is
there a way to sort efficiently this kind of big array, while leaving it
as much as possible on disk ?
I tried with TClonesArray but, apparently, it needs to be in main memory
to be sorted.
Any thoughts ? Any way to get that sorted information in the main Tree
and still have efficient access by time and not index ? I suppose this
kind of problem is common with "small" arrays.
Cheers
Damir
-- ===================================================================== | Damir Buskulic | Universite de Savoie/LAPP | | | Chemin de Bellevue, B.P. 110 | | Tel : +33 (0)450091600 | F-74941 Annecy-le-Vieux Cedex | | e-mail: buskulic@lapp.in2p3.fr | FRANCE | ===================================================================== mailto:buskulic@lapp.in2p3.fr