Some months ago there was discussions about 10 gbit performance with Linux. Some guys were pushing over 70 Gbit/sec through a single linux box.
70 Gbit/sec ? Maybe with port aggravation it's possible. Can you give some more info about that guys. To achieve that hight throughput maybe it's necessary to cut most of the OS and the kernel, leaving only the necessary. I'm very interested to read more information about the experiment.
regards
p.s here you can see 10 Gbit/s experiment http://haproxy.1wt.eu/10g.html
On Mon, Dec 21, 2009 at 12:04:32PM +0200, sadas sadas wrote:
Some months ago there was discussions about 10 gbit performance with Linux. Some guys were pushing over 70 Gbit/sec through a single linux box.
/centos@centos.org/pasik@iki.fi70 Gbit/sec ? Maybe with port aggravation it's possible. Can you give some more info about that guys. To achieve that hight throughput maybe it's necessary to cut most of the OS and the kernel, leaving only the necessary. I'm very interested to read more information about the experiment.
regards
p.s here you can see 10 Gbit/s experiment http://haproxy.1wt.eu/10g.html
See this thread: http://groups.google.com/group/linux.kernel/browse_thread/thread/70e62d8a85c...
quote: "We also achieved nearly 80 Gbps in bidirectional TCP tests (40 Gbps simultaneously in each direction):
This was using 2 dual-port 10-GigE NICs in the first two PCIe 2.0 slots. We are using an Intel i7 965 quad-core 3.2 GHz Nehalem processor (overclocked to 3.4 GHz) and 2000 MHz DDR3 memory. Adding an additional dual-port 10-GigE NIC on the Nvidia N200 chip does only marginally better, as it appears we are basically CPU limited at this point for this test (the sum of the TX and RX CPU utilization for each pair of 10-GigE interfaces is about 93%). "
-- Pasi