1 <!DOCTYPE HTML PUBLIC
"-//IETF//DTD HTML//EN">
5 <!-- WEBMAGIC VERSION NUMBER="2.0.1" -->
6 <!-- WEBMAGIC TRANSLATION NAME="ServerRoot" SRC="/var/www/htdocs/" DST="/" -->
7 <!-- WEBMAGIC TRANSLATION NAME="ProjectRoot" SRC="./" DST="" -->
8 <TITLE>What is Webstone
2.0</TITLE>
11 <CENTER><H1 ALIGN=
"CENTER"><IMG SRC=
"webstone.gif" WIDTH=
"534" HEIGHT=
"174" SGI_FULLPATH=
"/disk6/WebStone-2.0/doc/webstone.gif"></H1>
12 </CENTER><H1>Introducing WebStone
2.0</H1>
13 <P>WebStone
2.0 is the second generation Webstone web server benchmark. It
14 incorporates numerous bug fixes, modifications for compatibility with other
15 platforms and adds the new functionality of benchmark proxy servers, cgi
16 and NSAPI programs as well as introducing run rules which should make Webstone
17 numbers significantly more meaningful for comparison.
</P>
19 <P>Webstone
2.0 provides facilities for benchmarking proxy servers. This is
20 accomplished by putting in a value for the the PROXYSERVER entry in the
21 conf/testbed file, and changing the filelist to include URL's that have
22 the hostname for the actual web server.
</P>
23 <P>Dynamic content benchmarking is now explicitly supported in Webstone
2.0.
24 The file README.DynamicWorkload has directions for testing of NSAPI. The
25 included filelist.dynamic-{light,medium,heavy} serve as sample loads, with
26 the filelist.dynamic-heavy being the load that should be reported for NSAPI
27 performance. The cgi-send numbers should be quored for the filelist.cgi-heavy
29 <P>A port of the WebStone
2.0 benchmark to Windows NT is also included in this
30 release. This port is still in progress, so full functionality is not assured.
31 Specifically only the benchmark code has been ported - the supporting scripts
34 <P>As of Webstone
2.0, there are now run rules which must be adhered to for
35 published Webstone numbers. These are fairly basic, but they provide important
36 constraints on the benchmarking which make Webstone numbers more meaningful.
</P>
37 <P><B>Fileset:
</B>Included in the Webstone distribution is filelist.standard, which was previously
38 called filelist.sample. This filelist has a distribution of fileset sizes
39 that matches the kind of distributions seen in live web sites. The largest
40 file in the distribution is a
5 MB in length, which simulates the occasional
41 MPEG or other animation file which is downloaded. This filelist should be
42 used for all published Webstone numbers. Note that running WebStone
2.0
43 with the sort of fileset given in WebStone
1.1 will not yield a comparable
44 benchmark. In general, the WebStone
2.0 filelist will yield lower rates
45 for connections/second, but higher rates for throughput - the two sets of
46 numbers cannot be compared.
</P>
47 <P>When reporting NSAPI numbers, the filelist.dynamic-heavy filelist should
48 be used. For CGI numbers, the filelist.cgi-heavy filelist should be used.
</P>
49 <P><B>Benchmark Run Configuration:
</B> For a reported WebStone run, the runtime must be set at least
10 minutes.
50 This provides adequate time for the server and client configuration to reach
51 a steady state, and then provides a length of time long enough to cancel
52 out the high variations seen in the first few minutes of the run. The number
53 of clients should also vary from
20 to
100 in increments of
10 so that performance
54 of the server under a wide variety of loads can be observed.
</P>
55 <P><B>Server Configuration:
</B> The number of threads/processes is open to the discretion of the benchmarkers.
56 However, whether server side logging is on must be explicitly reported.
57 When logging is turned on, it must be in the common logfile format, and
58 only IP addresses should be logged. Parsed HTML is recommended to be turned
60 <P>Proxy Configuration: The configuration of how often the proxy server polls
61 the actual server for refreshes of it's cache should be described, as well
62 as any information
</P>
63 <P><B>Server Machine Configuration:
</B> When reporting runs, it is necessary that the operating system, memory
64 configuration and any special operating system modifications be reported,
65 especially changes to the TCP/IP stack.
</P>
66 <P><B>Testbed configuration:
</B>Reported runs must include information about the network topology being
67 used, as well as the number and type of machines generating load.
</P>
68 <P>All reported runs must include the information summarized by webstone -results,
69 excluding the timestamp. This includes: number of clients, connections per
70 second, little's law number, latency, error level and throughput. Preferably
71 in a table format.
</P>