BTW, since the blog post doesn't make it explicit: we're down to ~40 microseconds round trip between VMs in the same zone. No placement groups or infiniband required :).<p>Disclosure: I work on Google Cloud.
Why don't AWS people actively contribute in HN discussions about their achievements?<p>There're 8 comments on this thread and 3 of them are from GCP people giving less marketing-y insights. Thanks @jsolson and @boulos (it's always interesting to read your comments).
What prevents the hardware offloading from being used on public interfaces? Even if microsecond-level jitter reduction on public networks is negligible this should reduce CPU load in high PPS deployments, right?