Elephant Detection in the vSwitch With Performance Handling in the Underlay
Posted: May 16, 2014 Filed under: Uncategorized 2 Comments
As we’ve discussed previously, the vSwitch is a great position to detect elephant, or heavy-hitter flows because it has proximity to the guest OS and can use that position to gather additional context. This context may include the TSO send buffer, or even the guest TCP send buffer. Once an elephant is detected, it can be signaled to the underlay using standard interfaces such as DSCP. The following slide deck provides and overview of a working version of this, showing how such a setup can be used to both dynamically detect elephants and isolate mice from queuing delays they cause. We’ll write about this in more detail in a later post, but for now check out the slides (and in particular the graphs showing the latency of mice with and without detection and handling).
June 12, 2014 at 6:35 pm
Nice demonstration of the value of managing Elephant flows. However, there are situations where physical servers / appliances connect to the virtual network through a top of rack switch (implementing Network Services Gateway functionality), rather than the vSwitch.
Commodity switches are very capable devices and can be used for more that just tunnel termination. The following article demonstrates that a white box switch can also be used to detect and mark Elephant flows:
http://blog.sflow.com/2014/06/restful-control-of-cumulus-linux-acls.html
July 11, 2014 at 2:54 pm
[…] latter topic is something we’ve addressed in some other recent posts (here, here and here) — in this blog we’ll focus more on how we deal with physical devices at the […]