Questions related to Parallel Algorithms
I am looking for solving for the first few eigenvectors of a really large sparse symmetric matrix (1M x 1M). I am looking for suggestions as to which library will be best suited for my purpose. I am looking to compute the eigenvectors inside a few seconds. I have taken leverage of the library Spectra which uses the Iterative Arnoldi to solve the eigenvectors, but the time for computation has let me down. I have also looked into CUDA/MKL , but these libraries tend to take advantage of existing architectures. Is there some alternative or CUDA/MKL/OpenCL is the standard these days?
In our research institute, we are developping a new authomatic model selection technique using HPC in econometrics. Massive parallelism is applied to reduce selection algorithm' running times. To examine our code properties (speed up, latency, crashing probabilities, etc.) we will focus on alternative oil-price models. We will be very grateful if anyone could provide us with specific information about benchmarks or forecast comparisons on this subject.
How to convert Netcdf4 files to Netcdf3 files with NETCDF_nccopy
My system is Ubuntu 14.04, and netcdf-22.214.171.124has been installed
nccopy: nccopy [-k kind] [-[3|4|6|7]] [-d n] [-s] [-c chunkspec] [-u] [-w] [-[v|V] varlist] [-[g|G] grplist] [-m n] [-h n] [-e n] [-r] infile outfile
[-k kind] specify kind of netCDF format for output file, default same as input
kind strings: 'classic', '64-bit offset',
'netCDF-4', 'netCDF-4 classic model'
[-3] netCDF classic output (same as -k 'classic')
[-6] 64-bit-offset output (same as -k '64-bit offset')
[-4] netCDF-4 output (same as -k 'netCDF-4')
[-7] netCDF-4-classic output (same as -k 'netCDF-4 classic model')
[-d n] set output deflation compression level, default same as input (0=none 9=max)
[-s] add shuffle option to deflation compression
[-c chunkspec] specify chunking for dimensions, e.g. "dim1/N1,dim2/N2,..."
[-u] convert unlimited dimensions to fixed-size dimensions in output copy
[-w] write whole output file from diskless netCDF on close
[-v var1,...] include data for only listed variables, but definitions for all variables
[-V var1,...] include definitions and data for only listed variables
[-g grp1,...] include data for only variables in listed groups, but all definitions
[-G grp1,...] include definitions and data only for variables in listed groups
[-m n] set size in bytes of copy buffer, default is 5000000 bytes
[-h n] set size in bytes of chunk_cache for chunked variables
[-e n] set number of elements that chunk_cache can hold
[-r] read whole input file into diskless file on open (classic or 64-bit offset format only)
infile name of netCDF input file
outfile name for netCDF output file
netCDF library version 126.96.36.199 of Nov 6 2015 20:09:00 $
NetCDF: Unknown file format
Location: file nccopy.c; line 1354
Can anyone please suggest Partitioning algorithms to partition the vision algorithm (computations or workload) to expose opportunities for parallel execution by decomposing computations into small tasks
The system contains multi-processor with different types (Cpu, Gpu, and may contain soft-core)
I have found the paper on parallel version of rate monotonic algorithm. So it should be possible to get parallel version of DM/EDF. I want conformation form some knowledgeable person on the same to start the M.Tech project on same line.
The paper related to parallel version of rate monotonic algorithm is attached herewith.
MARE is a programming model and a run-me system that provides simple yet powerful abstractions for parallel, power-‐efficient software
– Simple C++ API allows developers to express concurrency
– User-‐level library that runs on any Android device, and on Linux, Mac OS X, and Windows platforms
I have noticed that CUDA is still prefered for parallel programming despite only be possible to run the code in a NVidia's graphis card. On the other hand, many programmers prefer to use OpenCL because it may be considered as a heterogeneous system and be used with GPUs or CPUs multicore. Then, I would like to know which one is your favorite and why?