Why such difference response times between Load Balancer and HAProxy?

Why is there seemingly such a big difference in response times between DigitalOcean’s Load Balancer (which as I gather uses HAProxy) and a $5 droplet with HAProxy installed? I ran several load tests on each pointing to the same thing and got these results:

Load Balancer: HAProxy:

Over a number of tests, the average response time was roughly the same around 775ms for Load Balancer and around 450ms for HAProxy. Is Load Balancer using an old version of HAProxy, on shared hosts or something else?

Submit an answer

This textbox defaults to using Markdown to format your answer.

You can type !ref in this text area to quickly search our full set of tutorials, documentation & marketplace offerings and insert the link!

Sign In or Sign Up to Answer

These answers are provided by our Community. If you find them useful, show some love by clicking the heart. If you run into issues leave a comment, or add your own answer to help others.

Accepted Answer

Hi there,

I’m the product manager for Load Balancers. Some of the parameters are not entirely clear or unknown (ie, the configuration of your HAProxy setup) , but there are some factors that could cause a discrepancy.

For example, if you are using the public static IP rather than a Floating IP, this could cause an additional latency, similar to what you are seeing, for each response since the LB product makes use of Floating IPs.

I should note that we are soon launching an update for Load Balancers which will include a performance increase, in addition to some new features. Be on the lookout for this!

Want to learn more? Join the DigitalOcean Community!

Join our DigitalOcean community of over a million developers for free! Get help and share knowledge in Q&A, subscribe to topics of interest, and get courses and tools that will help you grow as a developer and scale your project or business.