/[MITgcm]/MITgcm_contrib/llc_hires/llc_90/code-async/readme.txt
ViewVC logotype

Diff of /MITgcm_contrib/llc_hires/llc_90/code-async/readme.txt

Parent Directory Parent Directory | Revision Log Revision Log | View Revision Graph Revision Graph | View Patch Patch

revision 1.3 by dimitri, Sun Mar 3 01:17:30 2019 UTC revision 1.5 by dimitri, Wed Mar 6 16:54:57 2019 UTC
# Line 1  Line 1 
1  From Bron, February 28, 2019  From Bron Nelson, February 28, 2019
2    
3  In the newest version, it is no longer necessary to hand-edit the  In the newest version, it is no longer necessary to hand-edit the
4  constants in "recvTask.c" and "readtile_mpiio.c".  Instead, the file  constants in "recvTask.c" and "readtile_mpiio.c".  Instead, the file
# Line 10  information they need directly from "SIZ Line 10  information they need directly from "SIZ
10  for the run only need to be edited in one place (namely, SIZE.h).  for the run only need to be edited in one place (namely, SIZE.h).
11    
12  One tile per rank is recommended, mostly for pickup input performance,  One tile per rank is recommended, mostly for pickup input performance,
13  but it is not strictly necessary.  but it is not strictly necessary.  A minimum of one full node of I/O
14    ranks is required.  The async I/O does allocate whole nodes to be
15    either an I/O node, or a compute node.  It is permitted for the last
16    *compute* node to have a "ragged edge", i.e., have fewer MPI processes
17    on it than the other nodes do.  But the I/O nodes are all "full size".
18    
19    The other minimum value the I/O code requires is that there must
20    be at least one core for each field you want to write, e.g., if you
21    are dumping 20 different fields, there must be at least 20 cores
22    allocated to the I/O.  Note that the 20 (or whatever) number is
23    *aggregate* across all the I/O nodes, NOT a "per node" number.
24    
25    Another important constraint is that the total memory on all the I/O
26    nodes *collectively* needs to be twice as big as the largest epoch you
27    write.  So, if you are writing a 1.5 TB pickup dump, then you should
28    have a sum total of 3TB of memory (or more) on the set of I/O nodes.
29    
30  Choose dumpFreq and pChkptFreq as usual. We're not set up  Choose dumpFreq and pChkptFreq as usual. We're not set up
31  to do the rolling checkpoints yet. It'll dump u,v,t, and etan now -  to do the rolling checkpoints yet. It'll dump u,v,t, and etan now -
Legend:
Removed from v.1.3  
changed lines
  Added in v.1.5
  ViewVC Help
Powered by ViewVC 1.1.22