Update h2load howto and all manual pages

This commit is contained in:
Tatsuhiro Tsujikawa 2015-12-08 23:56:21 +09:00
parent 7cc2d22ab5
commit 86505b1c54
8 changed files with 75 additions and 24 deletions

View File

@ -1,6 +1,6 @@
.\" Man page generated from reStructuredText.
.
.TH "H2LOAD" "1" "November 29, 2015" "1.5.1-DEV" "nghttp2"
.TH "H2LOAD" "1" "December 08, 2015" "1.5.1-DEV" "nghttp2"
.SH NAME
h2load \- HTTP/2 benchmarking tool
.
@ -374,6 +374,28 @@ The fraction of the number of connections within standard
deviation range (mean +/\- sd) against total number of successful
connections.
.UNINDENT
.TP
.B req/s (client)
.INDENT 7.0
.TP
.B min
The minimum request per second among all clients.
.TP
.B max
The maximum request per second among all clients.
.TP
.B mean
The mean request per second among all clients.
.TP
.B sd
The standard deviation of request per second among all clients.
server.
.TP
.B +/\- sd
The fraction of the number of connections within standard
deviation range (mean +/\- sd) against total number of successful
connections.
.UNINDENT
.UNINDENT
.SH FLOW CONTROL
.sp

View File

@ -213,6 +213,8 @@ is 1 second and 500ms is 500 milliseconds). Units are h, m, s or ms
(hours, minutes, seconds and milliseconds, respectively). If a unit
is omitted, a second is used as unit.
.. _h2load-1-output:
OUTPUT
------
@ -301,6 +303,21 @@ time for 1st byte (of (decrypted in case of TLS) application data)
deviation range (mean +/- sd) against total number of successful
connections.
req/s (client)
min
The minimum request per second among all clients.
max
The maximum request per second among all clients.
mean
The mean request per second among all clients.
sd
The standard deviation of request per second among all clients.
server.
+/- sd
The fraction of the number of connections within standard
deviation range (mean +/- sd) against total number of successful
connections.
FLOW CONTROL
------------

View File

@ -1,3 +1,5 @@
.. _h2load-1-output:
OUTPUT
------

View File

@ -1,6 +1,6 @@
.\" Man page generated from reStructuredText.
.
.TH "NGHTTP" "1" "November 29, 2015" "1.5.1-DEV" "nghttp2"
.TH "NGHTTP" "1" "December 08, 2015" "1.5.1-DEV" "nghttp2"
.SH NAME
nghttp \- HTTP/2 client
.
@ -152,7 +152,8 @@ Default: \fB16\fP
.B \-M, \-\-peer\-max\-concurrent\-streams=<N>
Use <N> as SETTINGS_MAX_CONCURRENT_STREAMS value of
remote endpoint as if it is received in SETTINGS frame.
The default is large enough as it is seen as unlimited.
.sp
Default: \fB100\fP
.UNINDENT
.INDENT 0.0
.TP

View File

@ -116,7 +116,8 @@ OPTIONS
Use <N> as SETTINGS_MAX_CONCURRENT_STREAMS value of
remote endpoint as if it is received in SETTINGS frame.
The default is large enough as it is seen as unlimited.
Default: ``100``
.. option:: -c, --header-table-size=<SIZE>

View File

@ -1,6 +1,6 @@
.\" Man page generated from reStructuredText.
.
.TH "NGHTTPD" "1" "November 29, 2015" "1.5.1-DEV" "nghttp2"
.TH "NGHTTPD" "1" "December 08, 2015" "1.5.1-DEV" "nghttp2"
.SH NAME
nghttpd \- HTTP/2 server
.

View File

@ -1,6 +1,6 @@
.\" Man page generated from reStructuredText.
.
.TH "NGHTTPX" "1" "November 29, 2015" "1.5.1-DEV" "nghttp2"
.TH "NGHTTPX" "1" "December 08, 2015" "1.5.1-DEV" "nghttp2"
.SH NAME
nghttpx \- HTTP/2 proxy
.

View File

@ -1,10 +1,10 @@
h2load - HTTP/2 benchmarking tool - HOW-TO
==========================================
h2load is benchmarking tool for HTTP/2. If built with
h2load is benchmarking tool for HTTP/2 and HTTP/1.1. If built with
spdylay (http://tatsuhiro-t.github.io/spdylay/) library, it also
supports SPDY protocol. It supports SSL/TLS and clear text for both
HTTP/2 and SPDY.
supports SPDY protocol. It supports SSL/TLS and clear text for all
supported protocols.
Basic Usage
-----------
@ -22,29 +22,37 @@ In order to set benchmark settings, specify following 3 options.
If ``auto`` is given, the number of given URIs is used.
Default: ``auto``
For SSL/TLS connection, the protocol will be negotiated via ALPN/NPN.
You can set specific protocols in ``--npn-list`` option. For
cleartext connection, the default protocol is HTTP/2. To change the
protocol in cleartext connection, use ``--no-tls-proto`` option. For
convenience, ``--h1`` option forces HTTP/1.1 for both cleartext and
SSL/TLS connections.
Here is a command-line to perform benchmark to URI \https://localhost
using total 100000 requests, 100 concurrent clients and 10 max
concurrent streams::
concurrent streams:
.. code-block:: text
$ h2load -n100000 -c100 -m10 https://localhost
The benchmarking result looks like this::
The benchmarking result looks like this:
finished in 0 sec, 385 millisec and 851 microsec, 2591 req/s, 1689 kbytes/s
requests: 1000 total, 1000 started, 1000 done, 1000 succeeded, 0 failed, 0 errored
status codes: 1000 2xx, 0 3xx, 0 4xx, 0 5xx
traffic: 667500 bytes total, 28700 bytes headers, 612000 bytes data
.. code-block:: text
The number of ``failed`` is the number of requests returned with non
2xx status. The number of ``error`` is the number of ``failed`` plus
the number of requests which failed with connection error.
finished in 7.08s, 141164.80 req/s, 555.33MB/s
requests: 1000000 total, 1000000 started, 1000000 done, 1000000 succeeded, 0 failed, 0 errored, 0 timeout
status codes: 1000000 2xx, 0 3xx, 0 4xx, 0 5xx
traffic: 4125025824 bytes total, 11023424 bytes headers (space savings 93.07%), 4096000000 bytes data
min max mean sd +/- sd
time for request: 15.31ms 146.85ms 69.78ms 9.26ms 92.43%
time for connect: 1.08ms 25.04ms 10.71ms 9.80ms 64.00%
time to 1st byte: 25.36ms 184.96ms 79.11ms 53.97ms 78.00%
req/s (client) : 1412.04 1447.84 1426.52 10.57 63.00%
The number of ``total`` in ``traffic`` is the received application
data. If SSL/TLS is used, this number is calculated after decryption.
The number of ``headers`` is the sum of payload size of response
HEADERS (or SYN_REPLY for SPDY). This number comes before
decompressing header block. The number of ``data`` is the sum of
response body.
See the h2load manual page :ref:`h2load-1-output` section for the
explanation of the above numbers.
Flow Control
------------