Re:Making Facebook’s software infrastructure more energy eff...
by otterley on Saturday August 09, 2014 @11:39AM
(#8157104)
The Achilles heel of virtualization is networking. All of the hypervisors out there (VMWare, Xen, KVM) have user-space software switch implementations that dramatically reduce the throughput of TCP session creation. As a consequence you lose a significant amount of hardware potential to serve HTTP connections.
[ Parent
| on HN ]
Re:Making Facebook’s software infrastructure more energy eff...
by vicaya on Saturday August 09, 2014 @08:43PM
(#8158457)
This is wrong as well. VMware standard and distributed virtual switch is in hypervisor kernel.
Re:Making Facebook’s software infrastructure more energy eff...
by otterley on Sunday August 10, 2014 @11:28AM
(#8159942)
How's VMWare's TCP new-session-switching-rate compared to Xen's OVS? Any benchmarks showing a significant improvement?
Re:Making Facebook’s software infrastructure more energy eff...
by wmf on Saturday August 09, 2014 @09:12PM
(#8158517)
We just did some benchmarks on this topic: https://news.ycombinator.com/item?id=8146536
Re:Making Facebook’s software infrastructure more energy eff...
by reeze_xia on Saturday August 09, 2014 @12:17PM
(#8157194)
How about lightweight virtualization like linux container, docker?
Re:Making Facebook’s software infrastructure more energy eff...
by otterley on Saturday August 09, 2014 @02:17PM
(#8157560)
That's not virtualization; it's namespace isolation. There's a small performance impact if you're using NAT, but otherwise the kernel networking stack is used, so there's no performance penalty.
Re:Making Facebook’s software infrastructure more energy eff...
by reeze_xia on Sunday August 10, 2014 @10:46PM
(#8161707)
Yes, indeed. I mean the optimization of energy saving seems not suitable for the condition of namespace isolation, since you can't control the other containers requests. if we have to, we need to dispatch request from server perspective but not container.