Differences

This shows you the differences between two versions of the page.

--- library:computing:xios_impi_troubles [2017/08/04 14:26]
84.88.184.232 [Issue 3:]
+++ library:computing:xios_impi_troubles [2024/05/20 12:58]
84.88.52.107 old revision restored (2017/08/04 14:30)
@@ Line 202: / Line 202: @@
 **Solution:** Disabling compiler optimizations (activate -O0).
-==== Issue 3: ====
+==== Issue 3: MPI kills XIOS when writing model output ====
 **Environment:** Auto-EC-Earth 3.2.2_develop_MN4 (EC-Earth 3.2 r4063-runtime-unification).
@@ Line 259: / Line 259: @@
 We tried an alternative solution, which was to increment the number of XIOS servers in order to reduce the number of messages sent to the same process and by the moment it seems that it is effective.
-**Diagnosis:**
+**Diagnosis:** The problem is that an one point the buffer for data transfer is not 4-byte aligned, and the library assumes it is. This is a really low level problem and we do not completely understand the relation between this and reduce the congestion, but maybe in the future we can get more information.
-**Solution:**
+**Solution:** By the moment the solution used is to use enough number of XIOS servers (47 for SR).
 About Intel Communication Fabrics control:
 [[https://software.intel.com/en-us/node/528821]]
+Ips_proto.c source code:
+[[https://github.com/01org/psm/blob/master/ptl_ips/ips_proto.c]]
 ==== Issue 4: ====