Scheduled Downtime
On Tuesday 24 October 2023 @ 5pm MT the forums will be in read only mode in preparation for the downtime. On Wednesday 25 October 2023 @ 5am MT, this website will be down for maintenance and expected to return online later in the morning.
Normal Operations
The forums are back online with normal operations. If you notice any issues or errors related to the forums, please reach out to help@ucar.edu

box_rearrange::compute_dest:: ERROR

Mikasa

sky
Member
Hello, I am running compset B1PCTcmip6 with CESM2.1.3. But an error was reported :
box_rearrange::compute_dest:: ERROR: no destination found for compdof=
-2147427135
box_rearrange::compute_dest:: INFO: gsize= 320 384 18249
box_rearrange::compute_dest:: INFO: nioproc 12 ioproc 12
ioindex -1
box_rearrange::compute_dest:: INFO io 1 start=
1 1 1 count=
320 384 1520
And my pio setting are:
CPL : pio_numiotasks = 12
CPL : pio_stride = 64
CPL : pio_root = 1
CPL : pio_iotype = 6
The pelayout is :
./pelayout
Comp NTASKS NTHRDS ROOTPE
CPL : 768/ 1; 0
ATM : 768/ 1; 0
LND : 320/ 1; 0
ICE : 448/ 1; 320
OCN : 640/ 1; 768
ROF : 320/ 1; 0
GLC : 32/ 1; 0
WAV : 32/ 1; 1408
ESP : 1/ 1; 0
I have roughly determined the cause of the problem, that is, I add my code to read a so large nc file (about 122GB) in cpl that exceeds the memory limit. If I read a much smaller file, it can run successfully.
For some reason, I'd better read this huge file at once. So I wonder how can I make this ?
I think my HPC hardware resources are completely sufficient to do that with 8GB memory per core.

Thanks for your help.
 

jedwards

CSEG and Liaisons
Staff member
The message
compdof=-2147427135
indicates an integer overflow. Can the input file be divided into several files?
 

Mikasa

sky
Member
The message
compdof=-2147427135
indicates an integer overflow. Can the input file be divided into several files?
If I divide the file into 5 files, can I read them all in sequence ? That seems to cost the same amout of memory.
 

jedwards

CSEG and Liaisons
Staff member
memory is not the problem - the offset into the file is exceeding the limits of a 32bit integer. By splitting the file
you should be able to stay within the limits.
 
Top