On Sat, Sep 15, 2007 at 04:29:39PM -0700, David Schwartz wrote:I think he means that his application saturates the 10Gbps NIC with 8900 bytes datagrams and it consumes 100% of the CPU while doing this, and it's the same on the receiver side. I'm not surprized at all, considering that it means around 140kpps, or pauses of about 7 microseconds between each frame. Those pauses are too short to do anything else, resulting in the task to appear as taking 100% CPU. But IMHO, that does not mean that the machine is saturated; If the task had more work to do between each datagram, it would possibly be done. Same if it had to send shorter frames twice as fast. Willy -
| Jeremy Fitzhardinge | Re: [RFC 00/15] x86_64: Optimize percpu accesses |
| jmerkey | [ANNOUNCE] mdb: Merkey's Linux Kernel Debugger 2.6.27-rc4 released |
| Greg Kroah-Hartman | [PATCH 021/196] ISDN: Convert from class_device to device for ISDN capi |
| Ingo Molnar | Re: [PATCH 00/23] per device dirty throttling -v8 |
git: | |
| Linus Torvalds | Re: VCS comparison table |
| Peter Stahlir | Git as a filesystem |
| Johannes Schindelin | Re: git on MacOSX and files with decomposed utf-8 file names |
| Bill Lear | Meaning of "fatal: protocol error: bad line length character"? |
| Mayuresh Kathe | Re: What is our ultimate goal?? |
| Richard Stallman | Real men don't attack straw men |
| bofh | Re: web development on OpenBSD |
| Kevin | uvm_mapent_alloc: out of static map entries on 4.3 i386 |
| Mark Lord | Re: 2.6.25-rc8: FTP transfer errors |
| Evgeniy Polyakov | Re: [BUG] New Kernel Bugs |
| Jarek Poplawski | Re: [PATCH] pkt_sched: Destroy gen estimators under rtnl_lock(). |
| Andi Kleen | [PATCH RFC] [1/9] Core module symbol namespaces code and intro. |
