Philipp Kern <tr...@philkern.de> writes: > On 2010-02-20, Goswin von Brederlow <goswin-...@web.de> wrote: >> For network based boots, specifically high performance cluster, the size >> can make a real difference. When you turn the cluster on it is not just >> one system downloading an extra meg but 100+ nodes. That largely >> increases the network collisions, errors and dropped packages. Something >> that can even make systems fail to boot. > > Network collisions? That sounds so half duplex and wrong to me. Nowadays > we have switches. > > Kind regards, > Philipp Kern
When 100 nodes all want to talk to the one bootserver then that one poor port will be overflown. With switches you won't have collisions like in the old days when they would combine exponentially but you still get slowdowns. MfG Goswin -- To UNSUBSCRIBE, email to debian-devel-requ...@lists.debian.org with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org Archive: http://lists.debian.org/87pr3yua7f....@frosties.localdomain