As one of our feature enhancements we've improved the way our agents work, we are now consistently capturing key data on the actual API performance including:
- DNS Lookup time: i.e. 28355µs (28ms)
- Time to Connect: i.e 76106µs (76ms)
- Time for Handshake: i.e 0µs (0ms)
- Upload time: i.e. 48µs (0ms)
- Processing time: i.e. 120568µs (121ms)
- Download time: i.e. 2545µs (3ms)
However, this improvement has raised two issues. Firstly, we have realized that we had some minor reporting issues with our old collection agents which means that some of the latencies we are recording were actually better than actually were being experienced. Secondly, we saw a significant degrade in performance on the Softlayer agents with our new configuration - after investigation we found this to be due to a conflict between the way Softlayer's cloud service is handling IPv4 and IPv6 requests - leading to an extremely high percentage of DNS failures and retrys, which take approximately 5,000ms. We've worked with Softlayer and have a work around this now, but it's something to be aware of in the Cluster analysis in Insights, it's also something to be aware of if you use Softlayer as a hosting service. As ever, if you have any questions, please reach out to us.