Super User is a question and answer site for computer enthusiasts and power users. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I could not find this info any where else.

I am trying to figure out if the hackbench result on my embedded Linux device is good enough.

uname -a

Linux 2.6.26-rt #5 armv6l

Processor : ARMv6-compatible processor rev 3 (v6l)

RAM:256 MB
Nand: 3.0 GB

I am using

./hackbench -s 200 -l 1000 -g $i -f 30

where i ranges from 10-35. Data below:

Time for 10 process groups(secs) 196.246
Time for 15 process groups(secs) 289.823
Time for 20 process groups(secs) 390.367
Time for 25 process groups(secs) 492.254
Time for 30 process groups(secs) 580.849
Time for 35 process groups(secs) 682.971

Is this good performance ? Appreciate your feedback.

Thanks !

share|improve this question

migrated from Nov 21 '10 at 0:48

This question came from our site for professional and enthusiast programmers.

Good enough for what? – Matthew Flaschen Nov 20 '10 at 4:00
Indeed. Micro-benchmarks are useful for tracking changes made to a system and testing new code. They are not a substitute for testing against real use-cases and profiling the result. – stsquad Nov 20 '10 at 11:17
I have to say I disagree with the "off-topic" votes. This is a question in programming performance. As it stands it is underspecified enough to be NaRQ, and if fleshed out may be "too localized", but not "off-topic". That said it might (or might not) be better on E&R; at least, once more detail is provided. – dmckee Nov 21 '10 at 0:48

You must log in to answer this question.

Browse other questions tagged .