Changes between Version 3 and Version 4 of Public/User_Guide/TAMPI

Timestamp:

Jun 17, 2019, 3:15:06 PM (5 years ago)

Author:

Pedro Martinez-Ferror

Comment:

—

Public/User_Guide/TAMPI

@@ -183 +183 @@
 }}}
 
-
-
-
-----
+----
+
+
+= Nbody Benchmark (MPI+!OmpSs-2+TAMPI) =
+
+Users must clone/download this example's repository from [https://pm.bsc.es/gitlab/ompss-2/examples/nbody] and transfer it to a DEEP working directory.
+
+== Description ==
+
+This benchmark represents an N-body simulation to numerically approximate the evolution of a system of bodies in which each body continuously interacts with every other body.  A familiar example is an astrophysical simulation in which each body represents a galaxy or an individual star, and the bodies attract each other through the gravitational force.
+
+There are **7 implementations** of this benchmark which are compiled in different
+binaries by executing the command `make`. These versions can be blocking,
+when the particle space is divided into smaller blocks, or non-blocking, when
+it is not.
+
+The interoperability versions (MPI+!OmpSs-2+TAMPI) are compiled only if the environment variable `TAMPI_HOME` is set to the Task-Aware MPI (TAMPI) library's installation directory.
+
+== Execution Instructions ==
+
+The binaries accept several options. The most relevant options are the number
+of total particles (`-p`) and the number of timesteps (`-t`). More options
+can be seen with the `-h` option. An example of execution could be:
+
+`mpiexec -n 4 -bind-to hwthread:16 ./nbody -t 100 -p 8192`
+
+in which the application will perform 100 timesteps in 4 MPI processes with 16 hardware threads in each process (used by the !OmpSs-2 runtime). The total number of particles will be 8192 so that each process will have 2048 particles (2 blocks per process).
+
+== References ==
+
+* [https://pm.bsc.es/gitlab/ompss-2/examples/nbody]
+* [https://en.wikipedia.org/wiki/N-body_simulation]
+
+
+----
+
+
+= Heat Benchmark (MPI+!OmpSs-2+TAMPI) =
+
+Users must clone/download this example's repository from [https://pm.bsc.es/gitlab/ompss-2/examples/heat] and transfer it to a DEEP working directory.
+
+== Description ==
+
+This benchmark uses an iterative Gauss-Seidel method to solve the heat equation,
+which is a parabolic partial differential equation that describes the distribution of heat (or variation in temperature) in a given region over time. The heat equation is of fundamental importance in a wide range of science fields. In
+mathematics, it is the parabolic partial differential equation par excellence. In statistics, it is related to the study of the Brownian motion. Also, the diffusion equation is a generic version of the heat equation, and it is related to the study of chemical diffusion processes.
+
+There are **9 implementations** of this benchmark which are compiled in different
+binaries by executing the command `make`.
+
+The interoperability versions (MPI+!OmpSs-2+TAMPI) are compiled only if the environment variable `TAMPI_HOME` is set to the Task-Aware MPI (TAMPI) library's installation directory.
+
+== Execution Instructions ==
+
+The binaries accept several options. The most relevant options are the size
+of the matrix in each dimension (`-s`) and the number of timesteps (`-t`). More options can be seen with the `-h` option. An example of execution
+could be:
+
+`mpiexec -n 4 -bind-to hwthread:16 ./heat -t 150 -s 8192`
+
+in which the application will perform 150 timesteps in 4 MPI processes with 16
+hardware threads in each process (used by the !OmpSs-2 runtime). The size of the
+matrix in each dimension will be 8192 (8192^2^ elements in total), this means
+that each process will have 2048x8192 elements (16 blocks per process).
+
+== References ==
+
+* [https://pm.bsc.es/gitlab/ompss-2/examples/heat]
+* [https://pm.bsc.es/ftp/ompss-2/doc/examples/local/sphinx/04-mpi+ompss-2.html]
+* [https://en.wikipedia.org/wiki/Heat_equation]
+
+----
+
+= Krist Benchmark (!OmpSs-2+CUDA) =
+
+Users must clone/download this example's repository from [https://pm.bsc.es/gitlab/ompss-2/examples/krist] and transfer it to a DEEP working directory.
+
+== Description ==
+
+This benchmark represents the krist kernel, which is used in crystallography to find the exact shape of a molecule using Rntgen diffraction on single crystals or powders.
+
+There are **2 implementations** of this benchmark, ''krist'' and ''krist-unified'' using regular and unified CUDA memory, repectively.
+
+== Execution Instructions ==
+
+`./krist N_A N_R`
+
+where:
+* `N_A` is the number of atoms (1000 by default).
+* `N_R` is the umber of reflections (10000 by default).
+
+== References ==
+
+* [https://pm.bsc.es/gitlab/ompss-2/examples/krist]