Loading...
You are not logged-in Login/Register





  • Posts   Search Threads
  • matt.jNovember 13, 2009 1:35 AM PST   
    Interesting observation
    In an attempt to discern the maximum throughput of my [100Mbps] local network, and to a lesser extent see what level of overhead is introduced by the operating system (Windows), I performed a crude test which simply involved pushing as much traffic onto the link as possible destined to the next hop (router)

    Surprisingly, network utilization peaked at around 4MB/s (~33%); clearly very far from optimal.

    Long story short, the problem turned out to be the interrupt rate of the network adapter which was [during the initial test] set at 'adaptive', after modifying this to the lowest possible setting network utilization was able to maintain a stable ~99%. Oddly enough, Intel's recommendation was for higher interrupt rates when in high-load situations ... this actually produced a negative effect rendering < 12% utilization.


Forum jump:  

Intel Software Network Forums Statistics

16,374 users have contributed to 46,348 threads and 163,996 posts to date.

In the past 24 hours, we have 12 new thread(s) 83 new posts(s), and 58 new user(s).

In the past 3 days, the most popular thread for everyone has been Formula for the intersection of straight lines The most posts were made to Take a look at John Burkhard&# The post with the most views is \"-check none\" generates error

Please welcome our newest member shane122


For more complete information about compiler optimizations, see our Optimization Notice.