Web Server Benchmarks

Discussions center on performance benchmarks for web servers and frameworks, including requests for load testing results, throughput, latency, RPS metrics, and critiques of benchmarking tools and methodologies like wrk and HTTP/1.0.

📉 Falling 0.4x Web Development
3,101
Comments
20
Years Active
5
Top Authors
#7856
Topic ID

Activity Over Time

2007
6
2008
44
2009
90
2010
124
2011
126
2012
160
2013
267
2014
123
2015
182
2016
199
2017
192
2018
180
2019
152
2020
189
2021
257
2022
234
2023
203
2024
188
2025
172
2026
13

Keywords

e.g CPU loader.io RPS Loader.io ACT GP PER TPS VM load benchmark benchmarks throughput performance testing tests loaded http measuring

Sample Comments

leafboi Aug 14, 2020 View on HN

The tests aren't compute bound. They are testing requests per second. It's testing biased towards IO, not compute. Please read the article.

hn_1234 Jun 4, 2020 View on HN

Is there any load testing benchmark results available. if so can you share ?

panzerboiler Dec 7, 2022 View on HN

Those benchmarks are measuring the speed of the tool used to run the benchmark, not the http servers. Take them with a pack of salt.

miketheman Dec 29, 2021 View on HN

Good to know! Would you be interested in performing the same tests and using `wrk` and report the difference?

tericho Jul 8, 2013 View on HN

This is awesome, I'm wondering how it performs for the average web app. Are there benchmarks load tests anywhere?

brryant Apr 9, 2015 View on HN

Would love some throughput and other performance numbers!

cpach Jul 9, 2020 View on HN

Cool! What was the throughput and latency like?

zoombini29 May 4, 2016 View on HN

do you have perf benchmarks under heavier concurrency? (e.g., how does it do with 100 concurrent clients? 1000?)

segmondy Apr 3, 2019 View on HN

How did you benchmark it? rps? What was it doing in each request?

s3u May 24, 2011 View on HN

Any reason your tests used HTTP/1.0? It is likely that most of the time is spent in connection establishment and you may not have tested the server.