Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>Here's another benchmark (on Linux, just type <code>make</code>): <a href="http://dl.dropbox.com/u/5453551/blas_call_benchmark.zip" rel="noreferrer">http://dl.dropbox.com/u/5453551/blas_call_benchmark.zip</a></p> <p><a href="http://dl.dropbox.com/u/5453551/blas_call_benchmark.png" rel="noreferrer">http://dl.dropbox.com/u/5453551/blas_call_benchmark.png</a></p> <p>I do not see essentially any difference between the different methods for large matrices, between Numpy, Ctypes and Fortran. (Fortran instead of C++ --- and if this matters, your benchmark is probably broken.)</p> <p><s> Your <code>CalcTime</code> function in C++ seems to have a sign error. <code>... + ((double)start.tv_usec))</code> should be instead <code>... - ((double)start.tv_usec))</code>.</s> Perhaps your benchmark also has other bugs, e.g., comparing between different BLAS libraries, or different BLAS settings such as number of threads, or between real time and CPU time?</p> <p><strong>EDIT</strong>: failed to count the braces in the <code>CalcTime</code> function -- it's OK.</p> <p>As a guideline: if you do a benchmark, please always post <em>all</em> the code somewhere. Commenting on benchmarks, especially when surprising, without having the full code is usually not productive.</p> <hr> <p>To find out which BLAS Numpy is linked against, do:</p> <pre> $ python Python 2.7.2+ (default, Aug 16 2011, 07:24:41) [GCC 4.6.1] on linux2 Type "help", "copyright", "credits" or "license" for more information. >>> import numpy.core._dotblas >>> numpy.core._dotblas.__file__ '/usr/lib/pymodules/python2.7/numpy/core/_dotblas.so' >>> $ ldd /usr/lib/pymodules/python2.7/numpy/core/_dotblas.so linux-vdso.so.1 => (0x00007fff5ebff000) libblas.so.3gf => /usr/lib/libblas.so.3gf (0x00007fbe618b3000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fbe61514000) </pre> <p><strong>UPDATE</strong>: If you <em>can't</em> import numpy.core._dotblas, your Numpy is using its internal fallback copy of BLAS, which is slower, and not meant to be used in performance computing! The reply from @Woltan below indicates that this is the explanation for the difference he/she sees in Numpy vs. Ctypes+BLAS.</p> <p>To fix the situation, you need either ATLAS or MKL --- check these instructions: <a href="http://scipy.org/Installing_SciPy/Linux" rel="noreferrer">http://scipy.org/Installing_SciPy/Linux</a> Most Linux distributions ship with ATLAS, so the best option is to install their <code>libatlas-dev</code> package (name may vary).</p>
    singulars
    1. This table or related slice is empty.
    1. POBenchmarking (python vs. c++ using BLAS) and (numpy)
      singulars
      1. PTQuestion
    1. PTAnswer
    1. USpv.
    1. USpv.
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    3. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COI've run your benchmark; the results are [the same](http://i403.photobucket.com/albums/pp111/uber_ulrich/python-vs-cpp-blas/pv_plot.png)
      singulars
      1. PO
      1. USjfs
    2. COThank you very much for your posting. I ran your benchmark with [this](http://imageshack.us/photo/my-images/708/plothd.png/) result. So I cannot reproduce yours. To check which BLAS my numpy is using: I cannot `import numpy.core._dotblas`. What could be the problem here? I will try to clean up my benchmark an write a makefile in order for other to test it.
      singulars
      1. PO
      1. USWoltan
    3. CO@Woltan: the fact that you can't import numpy.core._dotblas means that your Numpy is using its internal fallback copy of BLAS (*slower*, and not meant to be used in performance computing!), rather than the BLAS library you have on your system. This explains the results you got from the benchmark. To fix the situation, you need to install a BLAS version Numpy can work with --- which means ATLAS or MKL. Here's a set of instructions: http://scipy.org/Installing_SciPy/Linux
      singulars
      1. PO
      1. USpv.
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload