Here's our first production Stack Overflow web server, before and after meltdown patching. Dammit. 
Here's another view. The peak impact is pretty severe over here:
Hardware info for the curious: 
The actual user impact is some spikes we need to now investigate. This is the 95th percentile view on response times for that same web server. Average went from 27ms to 29ms; less severe than the stall points. 
Here's what happened last night to one of our busier redis servers before and after patching for meltdown. Using the same 200k/12 minute axes on both: down from ~145k ops/sec to ~95ks ops/sec, or about 35% slower. 
That server is a Dell R630 running CentOS 7. It's dedicated to redis. Hardware details: 
This processor debacle fucking sucks. There's just no other way to put it. It has quite literally set us back YEARS of progress overnight.
