1

I am trying out a clustering tool developed by my senior and I don't have the source code. My problem is, I am unable to measure the memory and time requirement for a given clustering task.

I tried htop, but most of my clustering tasks run for only 1-2 seconds and thus, I am unable to monitor the process in htop. Further, if a clustering task takes reasonable amount of time (~40s), I am able to monitor the process in htop. However, it just shows 0.1% of MEM regardless of the input matrix file size (25MB and 58MB). How can I get a precise (or reasonably accurate) measurements.

I run the tool as follows:

./tool <clustering_algo_name> <input_matrix_file> <output_file>

I am running this on Ubuntu 12.04 (64 bit).

Maggie
  • 5,923
  • 8
  • 41
  • 56
  • 1
    See also http://stackoverflow.com/questions/131303/ – Matthew Strawbridge Apr 20 '14 at 08:47
  • "I am trying out a clustering tool developed by my senior and I don't have the source code" sounds like a horrible idea. Have you tried asking why source code isn't shared between colleagues working with the same stuff? –  Apr 21 '14 at 08:56

0 Answers0