Ragnar Lönn, the founder of Load Impact, presents his study on HTTP/2 performance. This slideshow was presented at Velocity NY, Oct. 13, 2015. Included in this presentation was the unveiling of a free application developers can use to benchmark their website performance on HTTP/1.1 vs HTTP/2 - http://http2.loadimpact.com/

1. HTTP/1.1 vs HTTP/2
Ragnar Lönn, Load Impact
ragnar@loadimpact.com
@ragnarlonn
- a performance analysis
Daniel Stenberg, Mozilla
daniel@haxx.se
@bagder

2. Part 2: The Experiment
Objective:
Try to get a sense of real-world
performance impact of going from
HTTP/1.1 to HTTP/2
@loadimpact

3. Approach
● Choose a well-known site
● Download all resources used by main page
● Host everything locally, in controlled
environment
● Measure load times while simulating
different network characteristics

4. Site: www.amazon.com
● ~230-240 resources to get to onload()
● ...but “only” ~10 javascript files
● ~10 unique source hosts
● ~7 Mbyte data

5. MacOS X 10.10.3 (4 CPU cores, 16G RAM)
VMware Fusion
Linux 3.19.0-25 (Ubuntu)
2 CPU cores, 2G RAM
● Nginx 1.9.5/http2
● Shimmer Cat 0.1
● Linux Netem
eth0:1
…
…
...
eth0:n
Chrome
The Setup

6. Test results

7. Test results
50-70% reduced load time
-66%

8. Remember that:
● HTTP/2 implementations still in their infancy
● Sites are optimized for HTTP/1.1
● Our lab setup is not 100% realistic
So who knows… But still, hey - 60%!

9. Interesting observations
Relative speed
change is flat

10. Dependencies

11. The dependency tree
index.html
style.css
script1.js
image1.png
image2.png
image3.png
script2.js
image4.png
image5.png
image6.png
image7.png

12. Dependency chains
index.html
style.css
script1.js
image1.png
image2.png
image3.png
script2.js
image4.png
image5.png
image6.png
image7.png

13. A dependency chain
index.html
script1.js
script2.js
image6.png
1
2
3
4

14. Dependency chain load time
Host1/index.html
Host1/script1.js
Host2/script2.js
Host3/image6.png
Conn Req
Req
Conn Req
Conn Req

15. The sum of all latencies
Host1/index.html
Host1/script1.js
Host2/script2.js
Host1 RTT Host2 RTT
Conn Req
Conn
Req
Host3 RTT
Host3/image6.png
Req
Conn Req

16. The largest sum of all latencies
index.html
style.css
script1.js
image1.png
image2.png
image3.png
script2.js
image4.png
image5.png
image6.png
image7.png

17. Minimum load time
Minimum theoretical load time =
● The largest “sum of all
latencies” value of all
dependency chains

18. Dependency chains
● # of resources loaded in one dependency
chain step exceeds # of TCP channels that
can be used concurrently
HTTP/1.1 is much disadvantaged when:
This happens quite often

19. Dependency chains
● The client needs to open new connections
HTTP/2 is disadvantaged when:
This happens more often than strictly necessary
today, as it often benefits HTTP/1.1 performance

20. Test @ http2.loadimpact.com