Comments (3)
This is a warning only, because we did not find advantage for
gathering more blocks into a larger buffer. The BG stands for the brigade
strategy, where this warning does not even make sense. The BG strategy in
MPI_AGGREGATE (the default strategy now), simply receives and writes out
each process' output data individually; there is no buffering of several
process' output together. The other, original strategy did buffer data from multiple
processes before writing a single large buffer; good
for small data/process type applications; the warning makes sense here,
although this strategy also works until you run out of memory.
We have removed this warning for the BG strategy and only give a detailed error message when there is not enough memory.
from adios.
Note, that you can choose the original aggregation strategy by passing the
'aggregation_type=1' argument to the MPI_AGGREGATE method (in the XML file, or in the adios_select_method() call). The BG strategy would be 2, which is now the default strategy.
<transport group="writer" method="MPI_AGGREGATE">
num_aggregators=64;num_ost=64;aggregation_type=1
</transport>
from adios.
Thanks for the info.
from adios.
Related Issues (20)
- ADIOS_READ_METHOD_BP_AGGREGATE not working in read API HOT 1
- bpdump crash on attribute file written by machine of different en HOT 4
- runconfig fail on theta@alcf HOT 3
- Numpy: Slicing is missing the stepping HOT 4
- python 3 support broken by print statements HOT 1
- SZ transformation build error HOT 2
- adios_mpi.cpp doesn't work with Python 3.7 HOT 2
- Support ZFP 0.5.1+ HOT 2
- Possible bug in bp2ncd utility HOT 1
- GFortran 10 Argument Mismatch HOT 2
- Question about timers HOT 4
- ADIOS Spack package broken on Summit HOT 5
- ZFP_INCLUDE_DIR not set HOT 1
- Compilation error HOT 6
- Attribute of Array of Complex
- Python DeprecationWarning: np.asscalar(a) is deprecated since NumPy v1.16, use a.item() instead HOT 1
- Python: wrapper of adios.File close errors
- Python 3.9 Build Error HOT 15
- Bug writing empty data sets? HOT 13
- Subsetting Data for in situ visualization HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from adios.