Cannot simulate 1E10 histories

Applications
1

Dear all,
I have run my simulation with 1E10 histories but It notified an error as shown in the following figure.
image
image

If I change to 5E9 histories, It simulated done without the error.
How can I fix the error? Thanks in advance.

2

The number of events is a G4int, which is a 32-bit signed integer.

1 Like
3

Dear,
I write it as a positive integer but it still gives the same error. But if it is written as 999999999, it will still run normally.

image
image751Ă—146 65 KB

image
image718Ă—554 167 KB

4

/run/beamOn accepts only integer values…and the maximum integer number is 2147483648 (i.e. 2^32/2)
if you need to simulate 1e10 events, you must use multiples runs

/run/beamOn n1
/run/beamOn n2
...

/run/beamOn n
1 Like
5

@faca87 could you please help me to generate event with multiple /run/beamOn and store data in a single root file , means the single root file must get updated for each run .

I followed your post " Set double for number of primary particles" but couldn’t succeed .

6

Just use in BeginOfRunAction

 if ( run->GetRunID() == 0 ) {
  analysisManager->OpenFile();
  }

and in the EndOfRunAction


  if ( run->GetRunID() == NUMBER) {
    analysisManager->Write();
    analysisManager->CloseFile();
  }

where NUMBER is your last run (example 10)

1 Like
7

hello @faca87
I have another query .

What was the typical size of the root file when you stored maximum event in a leaf ? For my case its 31 GB for 12*10^8 events .

8

It depends on a lot of things…what you are simulating, your setup, how many physical quantities you save,

For example, in my old simulation, I got 12-15GB root file when simulating 3*10^8 primary particles.
In a new one, I got 45-50GB root file when simulating 10^9 primary particles

1 Like