From be1c7e50e1e8809ea56f2c9d472eccd8ffd73a97 Mon Sep 17 00:00:00 2001 From: Daniel Baumann Date: Fri, 19 Apr 2024 04:57:58 +0200 Subject: Adding upstream version 1.44.3. Signed-off-by: Daniel Baumann --- web/server/h2o/libh2o/doc/benchmarks.html | 124 ++++++++++++++++++++++++++++++ 1 file changed, 124 insertions(+) create mode 100644 web/server/h2o/libh2o/doc/benchmarks.html (limited to 'web/server/h2o/libh2o/doc/benchmarks.html') diff --git a/web/server/h2o/libh2o/doc/benchmarks.html b/web/server/h2o/libh2o/doc/benchmarks.html new file mode 100644 index 00000000..2b90b80d --- /dev/null +++ b/web/server/h2o/libh2o/doc/benchmarks.html @@ -0,0 +1,124 @@ + + + + + + + + + + + + + + + +Benchmarks - H2O - the optimized HTTP/2 server + + +
+
+ +

+H2O +

+

the optimized HTTP/1.x, HTTP/2 server

+ + +
+ +
+ +
+
+
+Powered by Oktavia +
+
+ + +
+ + + + + + + + + + + + +
+ +

+Benchmarks +

+ + +

Download Timings

+ +
+

+Providing quick response to user is more important than anything else in web performance tuning. +According to a research conducted by Microsoft, 500msec slowdown in Bing causes their revenue go down by 1.2%1. +

+

+The chart below compares the first-paint times and download completion times of different web browsers / HTTP servers on a simulated network of 8Mbps bandwidth with 100ms latency, which is typical for today's mobile networks2. +

+
+ +
+

+It is clear in the case of this benchmark that the visitors of the web site would be more satisfied, if H2O was used as the HTTP server. +

+
+ +

Static-File Serving

+ +
+

+Below chart shows the scores recorded on Amazon EC2 running two c3.8xlarge instances (server and client) on a single network placement, serving a 612-byte file3. +For each measurement, 250 concurrent clients were used4. +open_file_cache was used for Nginx. +H2O implements a open-file-cache that gets updated immediately when the files are replaced. +

+
+ +
+
+ +

Reverse Proxy

+ +
+

+Presented below is an old chart showing the scores recorded on Amazon EC2 running two c3.8xlarge instances (server and client) on a single network placement56. +

+
+ +
+
+ + + +
+

Notes:

+
    +
  1. Velocity and the Bottom Line - O'Reilly Radar
    2. +
  2. A fork of http2rulez.com was used as the target website; bandwidth and latency were induced to local network using qdisc, specifically by running tc qdisc replace dev eth1 root handle 1:0 tbf rate 8192kbit burst 2048 latency 100ms; sudo tc qdisc add dev eth1 parent 1:1 netem delay 100ms, and sysctl -w net.ipv4.tcp_no_metrics_save=1.
    3. +
  3. Configuration files used: nginx.conf, h2o.conf.
    4. +
  4. Wrk was used for HTTP/1 tests. h2load was used for HTTP/2.
    5. +
  5. For reverse-proxy tests, another H2O process running on the same host was used as the upstream server
    6. +
  6. open-file-cache was not used in the static-file benchmark
    7. +
+
+ +
+ + + -- cgit v1.2.3