Message: thread merge problem Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Question thread merge problem 

Forum: Multithreading
Date: 20 Jan, 2017
From: Huan Yao <Huan Yao>

Dear Geant4 experts,
My system is
Ubuntu 16.04 in VirtualBox (windows 10)
gcc 5.4
geant4 10.3

I use G4RootAnalsisManager to save vector<int> "i" and vector<double> "x" to the root, for each entry, the size of those two variables are same.
I didn't find any problem until I check the size of the variable for each entry with root. They are different, but total entries=entries*size are same. It seems to me Geant4 is merging but with random order.

After a few trials, I found it's because of "SetNtupleMerging" option. Once it's true, the size is different for some entry, not all. When it's off, I use hadd to merge threaded root files, test_t*.root. Everything is right.

I also test with G4MPI on/off, another test on linux server, same situation.

Any suggestion about this SetNtupleMerging option? Do I need to do sth to make sure it's in the order?

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 None: Re: thread merge problem   (Ivana Hrivnacova - 23 Jan, 2017)
(_ None: Re: thread merge problem   (Huan Yao - 23 Jan, 2017)
(_ None: Re: thread merge problem   (Ivana Hrivnacova - 23 Jan, 2017)
(_ None: Re: thread merge problem   (Huan Yao - 27 Jan, 2017)
(_ None: Re: thread merge problem   (Ivana Hrivnacova - 07 Feb, 2017)
(_ None: Re: thread merge problem   (Huan Yao - 12 Feb, 2017)
(_ None: Re: thread merge problem   (Ivana Hrivnacova - 23 Feb, 2017)
 Add Message Add Message
to: "thread merge problem"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews