[Opendnssec-develop] hsmspeed not optimal?

Rick Zijlker rick.zijlker at sidn.nl
Thu Jan 28 16:09:06 CET 2010


Hey all,

 

I'm running hsmspeed with 8 threads and lots of iterations, but it looks
like it's not using the full potential of the machine. The load gets
evenly divided over the 8 cores.

 

Is this intended? This way I can't get an estimate of the performance
when signing with 8 full cores.

 

PRC | sys   0.38s | user   9.79s | #thr     149 | #zombie    1 | #exit
0 |

CPU | sys      4% | user     98% | irq       0% | idle    698% | wait
0% |

cpu | sys      0% | user     13% | irq       0% | idle     86% | cpu000
w  0% |

cpu | sys      1% | user     13% | irq       0% | idle     86% | cpu006
w  0% |

cpu | sys      1% | user     13% | irq       0% | idle     86% | cpu004
w  0% |

cpu | sys      0% | user     13% | irq       0% | idle     87% | cpu007
w  0% |

cpu | sys      1% | user     12% | irq       0% | idle     87% | cpu005
w  0% |

cpu | sys      1% | user     12% | irq       0% | idle     87% | cpu002
w  0% |

cpu | sys      0% | user     12% | irq       0% | idle     88% | cpu001
w  0% |

cpu | sys      0% | user      9% | irq       0% | idle     90% | cpu003
w  0% |

MEM | tot   15.7G | free   10.9G | cache   4.0G | buff  409.3M | slab
224.4M |

SWP | tot   16.0G | free   16.0G |              | vmcom 273.1M | vmlim
23.8G |

DSK |  cciss/c0d0 | busy      0% | read       0 | write      6 | avio
4 ms |

NET |   transport | tcpi       1 | tcpo       1 | udpi       0 | udpo
0 |

NET |     network | ipi      201 | ipo        1 | ipfrw      0 | deliv
201 |

NET | dev    eth0 | pcki     201 | pcko       1 | in   28 Kbps | out   0
Kbps |

 

  PID  SYSCPU  USRCPU  VGROW  RGROW USERNAME THR  ST EXC  S  CPU CMD
1/1

19701   0.37s   9.78s     0K     8K root       9  --   -  S 102%
ods-hsmspeed

20700   0.01s   0.01s     0K     0K root       1  --   -  R   0% atop

 

Cheers,

Rick

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.opendnssec.org/pipermail/opendnssec-develop/attachments/20100128/27eb5c46/attachment.html>


More information about the Opendnssec-develop mailing list