9%
25.02.2013
If you are a system administrator of many systems, or even of just a desktop or laptop, you are likely monitoring your system in some fashion. This is particularly true in high-performance computing
9%
12.02.2013
of Open MPI installed and a user wanted to try the PetSc libraries with a new version, you could easily install and build everything in /opt
and have the user running new code without rebooting nodes
11%
30.01.2013
as well), but you might also have users who need previous versions of these packages. This problem is compounded by having multiple compilers and multiple MPI libraries, resulting in a large number
9%
16.01.2013
method to calculate the value of pi:
$ grep -v local /etc/hosts | cut -d" " -f2 > ~/hostfile
$ nano pi.py
$ mpirun -np 2 -hostfile hostfile python pi.py
3.14192133333
Listing 8: pi.py
01 from mpi4py
9%
03.01.2013
-ons, such as MPI, and rewriting the code. This approach allows you to start multiple instances of the tool on different nodes and have them communicate over a network so that code can be executed in parallel.
I won
10%
19.12.2012
’t cover it here.
MPI Profiling and Tracing
For HPC, it’s appropriate to discuss how to profile and trace MPI (Message-Passing Interface) applications. A number of MPI profiling tools are available
35%
04.12.2012
was particularly effective in HPC because clusters were composed of singe- or dual-processor (one- or two-core) nodes and a high-speed interconnect. The Message-Passing Interface (MPI) mapped efficiently onto ... HPC, parallel processing, GPU, multicore, OpenMP, MPI, many core, OpenACC, CUDA, MICs, GP-GPU
10%
24.11.2012
+ command-line interface. It includes updates to many modules, including: the HPC Roll (which contains a preconfigured OpenMPI environment), as well as the Intel, Dell, Univa Grid Engine, Moab, Mellanox, Open
10%
21.11.2012
, it’s very easy to get laptops with at least two, if not four, cores. Desktops can easily have eight cores with lots of memory. You can also get x86 servers with 64 cores that access all of the memory
11%
06.11.2012
options, and you notice that some simple options are a choice of MPI and BLAS libraries. Of course, you also need to choose a compiler. The task seems simple enough until you lay out the possible choices