1

I am using Sugar (a SAT-based Constraint Solver) to run an experiment, for each new experiment I increase the size of the sugar file, which means that the percent of CPU and memory usage should increase. However, the CPU percentage is lower for larger files which don't make sense, can anyone help me to explain this behavior? the specification of the Sugar and the server used t run the experiment is shown in the image. machne specifications used in expirement

Alaa Oqaily
  • 11
  • 1
  • Do not provide the details as a link. Post them as a part of your question. – Jeroen Heier Dec 26 '18 at 16:11
  • Have you tried measuring the number of [*cache-miss*, *TLB miss* and *page faults*](https://stackoverflow.com/questions/37825859/cache-miss-a-tlb-miss-and-page-fault) on your system as you increase the size of the input, and compare them with the decrease in CPU usage? – Patrick Trentin Jan 04 '19 at 10:18
  • Patrick Trentin Thank you, the CPU decreases due to the page fault indeed. can you help me with a way to minimize the number of page faults? – Alaa Oqaily Feb 06 '19 at 02:11

0 Answers0