Re: [ovs-discuss] Performance of short-lived TCP sessions

2014-02-03 Thread McGarvey, Kevin
No problem, I figured you meant that. On 2/3/14 3:34 PM, "Justin Pettit" wrote: >On 2/3/14, 12:30 PM, "Justin Pettit" wrote: > >>On 2/3/14, 12:17 PM, "McGarvey, Kevin" wrote: >> >>>The TCP workload of a global DNS server consists of T

Re: [ovs-discuss] Performance of short-lived TCP sessions

2014-02-03 Thread McGarvey, Kevin
That's good news. I'll test it. Kevin On 2/3/14 3:30 PM, "Justin Pettit" wrote: >On 2/3/14, 12:17 PM, "McGarvey, Kevin" wrote: > >>The TCP workload of a global DNS server consists of TCP sessions, each >>with a single request and a response th

[ovs-discuss] Performance of short-lived TCP sessions

2014-02-03 Thread McGarvey, Kevin
The TCP workload of a global DNS server consists of TCP sessions, each with a single request and a response that is typically under 2KB, and there is a very large number of source IPs. What level of performance, expressed in terms of sessions per second, would be expected from Open vSwitch with

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-24 Thread McGarvey, Kevin
On 1/24/14 8:36 AM, "Flavio Leitner" wrote: >On Wed, Jan 22, 2014 at 09:40:11PM +0000, McGarvey, Kevin wrote: >> >> >> On 1/22/14 4:10 PM, "Ben Pfaff" wrote: >> >> >On Wed, Jan 22, 2014 at 09:04:48PM +, McGarvey, Kevin wrote: &g

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
On 1/22/14 6:30 PM, "Gurucharan Shetty" wrote: >> >> Where can I find that? I installed OVS from an rpm. I ran rpm -qlp on >> the package and I didn't see the kernel module. Do I need to build it? > >Where did you get your RPMS from? Did you build it? > >RHEL comes with a default OVS kernel m

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
On 1/22/14 4:42 PM, "Ben Pfaff" wrote: >On Wed, Jan 22, 2014 at 09:40:11PM +0000, McGarvey, Kevin wrote: >> >> >> On 1/22/14 4:10 PM, "Ben Pfaff" wrote: >> >> >On Wed, Jan 22, 2014 at 09:04:48PM +, McGarvey, Kevin wrote: &g

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
On 1/22/14 4:10 PM, "Ben Pfaff" wrote: >On Wed, Jan 22, 2014 at 09:04:48PM +0000, McGarvey, Kevin wrote: >> >> >> On 1/22/14 3:23 PM, "Ben Pfaff" wrote: >> >> >On Wed, Jan 22, 2014 at 08:17:05PM +, McGarvey, Kevin wrote: >

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
On 1/22/14 3:23 PM, "Ben Pfaff" wrote: >On Wed, Jan 22, 2014 at 08:17:05PM +0000, McGarvey, Kevin wrote: >> >> >> On 1/22/14 12:44 PM, "Ben Pfaff" wrote: >> >> >On Wed, Jan 22, 2014 at 09:39:14AM -0800, Ben Pfaff wrote: >> >

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
On 1/22/14 12:44 PM, "Ben Pfaff" wrote: >On Wed, Jan 22, 2014 at 09:39:14AM -0800, Ben Pfaff wrote: >> On Wed, Jan 22, 2014 at 05:35:40PM +0000, McGarvey, Kevin wrote: >> > >> > >> > On 1/21/14 6:17 PM, "Ben Pfaff" wrote: >>

Re: [ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-22 Thread McGarvey, Kevin
ance I should see? > >Version 2.0 helped with this more, too. > >On Tue, Jan 21, 2014 at 11:14:03PM +, McGarvey, Kevin wrote: >> Sorry, I left out details in order to get an answer about whether kernel >> forwarding of single packet flows like this is even possibl

[ovs-discuss] Open vSwitch performance with UDP traffic

2014-01-21 Thread McGarvey, Kevin
I am sending UDP DNS queries to a KVM guest through Open vSwitch running on the RHEL hypervisor. With traffic of only 5K request/response pairs per second the CPU consumption of the ovs-vswitchd process is over 60%. The source IP and port change with each request, simulating normal UDP DNS traf