Skip to content

Commit cf11a95

Browse files
ChrisJ60ChrisJ60
committed
Added support for Solaris 11. many other improvements.
This version and later will NOT inter-operate with versions prior to this one.
1 parent 5ccc8af commit cf11a95

15 files changed

+1244
-211
lines changed

‎NwtestHelp.txt ‎HelpLinux.txt

+17-6
Original file line numberDiff line numberDiff line change
@@ -1,5 +1,5 @@
11

2-
Version 2.2
2+
Version 2.3
33

44
This program implements a network response time and throughput test.
55

@@ -31,8 +31,10 @@ number of connections.
3131

3232
Usage:
3333

34-
nwtest h[elp] { h[elp] | u[sage] | g[eneral] | c[lient] |
35-
s[erver] | m[etrics] | f[ull] }
34+
nwtest h[elp] { h[elp] | i[nfo] | u[sage] | g[eneral] |
35+
c[lient] | s[erver] | m[etrics] | f[ull] }
36+
37+
nwtest i[nfo]
3638

3739
nwtest s[erver] <port> [-4|-6] [-h[ost] <h>] [-m[sgsz] <m>]
3840
[-c[onn] <c>] [-l[og] <logpath>]
@@ -50,6 +52,11 @@ nwtest h[elp] { u[sage] | g[eneral] | c[lient] | s[erver] | m[etrics] | f[ull] }
5052
Display help on the specified topic.
5153

5254

55+
nwtest i[nfo] }
56+
57+
Display information about this utility.
58+
59+
5360
nwtest s[erver] <port> [-4|-6] [-h[ost] <h>] [-m[sgsz] <m>]
5461
[-c[onn] <c>] [-l[og] <logpath>]
5562

@@ -128,15 +135,19 @@ client and server), '-rbsz' (sets size for the receive buffer on both client
128135
and server), '-srvsbsz' (sets size for send buffer on server), '-srvrbsz'
129136
(sets size for receive buffer on server), '-cltsbsz' (sets size for send
130137
buffer on client) and '-cltrbsz' (sets size for receive buffer on client).
131-
These values are specified in bytes and each must be between 8,192 and
132-
8,388,608. Also, the total of the sizes must be <= 8,388,608. These sizes may
138+
These values are specified in bytes and each must be between 4,096 and
139+
4,194,304. Also, the total of the sizes must be <= 4,194,304. These sizes may
133140
be specified using a suffix of 'k' to repesent KB (1,024 bytes) or 'm' to
134141
represent MB (1,048,576 bytes).
135142

143+
Other factors, such as OS configuration, may place further constraints on
144+
the maximum allowed send and receive buffer sizes. Use 'nwtest info' to
145+
get more specific information.
146+
136147
If '-nodelay' is specified then the TCP_NODELAY option is enabled on all
137148
sockets used for data transfer in both the client and the server.
138149

139-
If '-quickack' is specified then the TCP_SENDMOREACKS option is enabled on
150+
If '-quickack' is specified then the TCP_QUICKACK option is enabled on
140151
all sockets used for data transfer in both the client and the server.
141152

142153
Normally only aggregate performance metrics are displayed, but if '-verbose'

‎HelpMacOS.txt

+296
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,296 @@
1+
2+
Version 2.3
3+
4+
This program implements a network response time and throughput test.
5+
6+
The program can be run as either a client or a server. The client opens one
7+
or more connections to a server and exchanges messages of a given size with
8+
the server for a given period of time. At the end of the test period, various
9+
performance metrics are displayed.
10+
11+
The test can run in two modes. In synchronous (request/response) mode, each
12+
connection has a single thread in both the client and the server. Once the
13+
connection is established the client sends a message to the server and waits
14+
for a response. As soon as the server receives a message it sends it straight
15+
back to the client. This sequence is repeated until the test ends.
16+
17+
In asynchronous (streaming) mode, each connection has two threads in both the
18+
client and the server. Once the connection is established the client sends a
19+
continuous stream of messages to the server and the server simultaneously
20+
sends a continuous stream of messages to the client.
21+
22+
The size of the messages exchanged, the duration of the test measurement phase,
23+
the amount of load ramp-up and ramp-down time and the number of connections
24+
(and hence threads) are all configurable.
25+
26+
The defaults for the various configurable parameters have been carefully chosen
27+
such that using the defaults will generally give a meaningful result. The main
28+
parameters that you might vary in normal usage are the message size and the
29+
number of connections.
30+
31+
32+
Usage:
33+
34+
nwtest h[elp] { h[elp] | i[nfo] | u[sage] | g[eneral] |
35+
c[lient] | s[erver] | m[etrics] | f[ull] }
36+
37+
nwtest i[nfo]
38+
39+
nwtest s[erver] <port> [-4|-6] [-h[ost] <h>] [-m[sgsz] <m>]
40+
[-c[onn] <c>] [-l[og] <logpath>]
41+
42+
nwtest c[lient] <host> <port> [-s[rc] <srcaddr>] [-4|-6] [-a[sync]]
43+
[-c[onn] <c>] [-d[ur] <d>] [-r[amp] <r>]
44+
[-m[sgsz] <m>] [-l[og] <logpath>] [-e[cn]]
45+
[-sbsz <sbsz> | [[-srvsbsz <srvsbsz>] [-cltsbsz <cltsbsz>]]
46+
[-rbsz <rbsz> | [[-srvrbsz <srvrbsz>] [-cltrbsz <cltrbsz>]]
47+
[-n[odelay]] [-q[uickack]] [-b[rief]|-v[erbose]]
48+
49+
50+
nwtest h[elp] { u[sage] | g[eneral] | c[lient] | s[erver] | m[etrics] | f[ull] }
51+
52+
Display help on the specified topic.
53+
54+
55+
nwtest i[nfo] }
56+
57+
Display information about this utility.
58+
59+
60+
nwtest s[erver] <port> [-4|-6] [-h[ost] <h>] [-m[sgsz] <m>]
61+
[-c[onn] <c>] [-l[og] <logpath>]
62+
63+
Run as a server on local port <port>. If a specific host is specified (<h>)
64+
then bind to the address(es) for that host, otherwise bind to INADDR[6]_ANY.
65+
The host can be specified as a hostname or an IP address (IPv4 or IPv6);
66+
the address specified must be an address for an interface on the local system.
67+
If a hostname is specified then you can use the '-4' or '-6' options to limit
68+
communication to IPv4 or IPv6.
69+
70+
The maximum message size, in bytes, that the server will accept is specified
71+
by <m> where 28 <= <m> <= 1,048,576. Connections requesting a message size
72+
larger than this will be rejected. The default is 1,048,576. This size
73+
represents the size of the user data; it excludes network protocol overheads
74+
(ethernet, IPv4/6 etc.). The size may be specified using a suffix of 'k' to
75+
repesent KB (1,024 bytes) or 'm' to represent MB (1,048,576 bytes).
76+
77+
The maximum number of concurrent connections that the server will allow is set
78+
using <c> where 2 <= <c> <= 128. Connections that exceed this number will
79+
be rejected. The default is 32.
80+
81+
Normally all output goes to stdout/stderr, but if '-log' is specified then
82+
after initial argument parsing any subsequent messages will be written only
83+
to <logpath> with microsecond resolution timestamps. A <logpath> of '-'
84+
equates to 'stdout' and '--' equates to 'stderr'.
85+
86+
87+
nwtest c[lient] <host> <port> [-s[rc] <srcaddr>] [-4|-6] [-a[sync]]
88+
[-c[onn] <c>] [-d[ur] <d>] [-r[amp] <r>]
89+
[-m[sgsz] <m>] [-l[og] <logpath>] [-e[cn]]
90+
[-sbsz <sbsz> | [[-srvsbsz <srvsbsz>] [-cltsbsz <cltsbsz>]]
91+
[-rbsz <rbsz> | [[-srvrbsz <srvrbsz>] [-cltrbsz <cltrbsz>]]
92+
[-n[odelay]] [-q[uickack]] [-b[rief]|-v[erbose]]
93+
94+
Run as a client connecting to a server at host <host> and port <port>. The
95+
host can be specified as either a host name or an IP address (IPv4 or IPv6).
96+
If a hostname is specified then you can use the '-4' or '-6' options to limit
97+
communication to IPv4 or IPv6.
98+
99+
Normally the OS will determine the local source IP address (interface) to use
100+
for the outgoing connection, but this can be overridden by specifying an
101+
explicit local address using the '-src' option. This value must be an IP
102+
address, not a hostname, and it must correspond to the address of an
103+
interface on the local system. If <host> has been specified using an IP
104+
address then the address type (IPv4 or IPv6) for <srcaddr> must be the same
105+
as that of <host>. Furthermore, if the allowed connection type has been
106+
restricted by way of '-4' or '-6' then the value of <srcaddr> must be of
107+
the selected type.
108+
109+
By default the test is performed in request/response (synchronous) mode but
110+
if '-async' is specified the test is performed in streaming mode.
111+
112+
The measurement part of the test will run for <d> seconds with a ramp-up/down
113+
of <r> seconds. 10 <= <d> <= 300 with a default of 30 and 0 <= <r> <= 30
114+
with a default of 10.
115+
116+
The message size used for the test is specified by <m> where 28 <= <m> <=
117+
1,048,576 with a default of 1,024. This size represents the size of the user
118+
data; it excludes network protocol overheads (ethernet, IPv4/6 etc.). The
119+
size may be specified using a suffix of 'k' to repesent KB (1,024 bytes) or
120+
'm' to represent MB (1,048,576 bytes).
121+
122+
The number of concurrent connections used is specified by <c> where 1 <= <c>
123+
<= 64 with a default of 1.
124+
125+
Normally all output goes to stdout/stderr, but if '-log' is specified then
126+
after initial argument parsing any subsequent messages will be written only
127+
to <logpath> with microsecond resolution timestamps. A <logpath> of '-'
128+
equates to 'stdout' and '--' equates to 'stderr'.
129+
130+
If you specify '-ecn', Explicit Congestion Notification is requested for
131+
the connection. The system will attempt to negotiate ECN but there is no
132+
guarantee that it will succeed and no indication if the negotiation fails.
133+
134+
Normally the OS will allocate the sizes for the socket send and receive
135+
buffers, and these sizes will be reported in the connection messages. If
136+
you want to use specific values for the socket send and receive buffer
137+
sizes then you can do so using '-sbsz' (sets size for send buffer on both
138+
client and server), '-rbsz' (sets size for the receive buffer on both client
139+
and server), '-srvsbsz' (sets size for send buffer on server), '-srvrbsz'
140+
(sets size for receive buffer on server), '-cltsbsz' (sets size for send
141+
buffer on client) and '-cltrbsz' (sets size for receive buffer on client).
142+
These values are specified in bytes and each must be between 4,096 and
143+
4,194,304. Also, the total of the sizes must be <= 4,194,304. These sizes may
144+
be specified using a suffix of 'k' to repesent KB (1,024 bytes) or 'm' to
145+
represent MB (1,048,576 bytes).
146+
147+
Other factors, such as OS configuration, may place further constraints on
148+
the maximum allowed send and receive buffer sizes. Use 'nwtest info' to
149+
get more specific information.
150+
151+
If '-nodelay' is specified then the TCP_NODELAY option is enabled on all
152+
sockets used for data transfer in both the client and the server.
153+
154+
If '-quickack' is specified then the TCP_SENDMOREACKS option is enabled on
155+
all sockets used for data transfer in both the client and the server.
156+
157+
Normally only aggregate performance metrics are displayed, but if '-verbose'
158+
is specified then per connection metrics are also displayed. If '-brief' is
159+
specified then just key metrics are displayed on a single line.
160+
161+
162+
For each connection successfully established, the client and server will
163+
report the TCP MSS (maxseg) value and the size of the socket send and
164+
receive buffers (sndbsz anb rcvbsz).
165+
166+
The metrics measured and reported by this program for each test mode are
167+
as follows; all references to 'data' and 'throughput' refer to application
168+
data excluding network overheads.
169+
170+
All modes
171+
---------
172+
173+
Elapsed time - The wall clock elapsed time for the measurement part
174+
of the test (excludes ramp up/down time).
175+
176+
User CPU time - The amount of user CPU time consumed during the
177+
elapsed time.
178+
179+
System CPU time - The amount of system CPU time consumed during the
180+
elapsed time.
181+
182+
Total CPU time - User time plus system time.
183+
184+
Process CPU usage - The average CPU usage for the nwtest process during
185+
the elapsed time, expressed as a percentage of one
186+
CPU core.
187+
188+
System CPU usage - The average CPU usage for the nwtest process during
189+
the elapsed time, expressed as a percentage of total
190+
available system CPU resources.
191+
192+
Sync (request/response) mode
193+
----------------------------
194+
195+
Total messages - The total number of data messages sent during the
196+
measurement period. The number of received messages
197+
is the same.
198+
199+
Total data - The total number of bytes sent during the measurement
200+
period. The number of received bytes is the same.
201+
202+
Avg measure time - The average measurement time across all threads (µs).
203+
204+
Start variance - The maximum difference between the start times of all
205+
the threads (µs). Only displayed if connections > 1.
206+
207+
Run variance - The maximum difference between the measurement times of
208+
all the threads (µs). Only displayed if connections > 1.
209+
210+
Throughput - The send throughput, aggregated across all connections,
211+
during the measurement period (bytes/second).
212+
213+
Minimum R/T - The lowest round trip time across all connections during
214+
the measurement period (µs).
215+
216+
Average R/T - The average round trip time across all connections during
217+
the measurement period (µs).
218+
219+
Maximum R/T - The highest round trip time across all connections during
220+
the measurement period (µs).
221+
222+
In brief mode the output consists of a single line as follows:
223+
224+
info: results S,<nconn>,<throughput>,<minrt>,<avgrt>,<maxrt>,<proccpu>,<syscpu>
225+
226+
Async (streaming) mode
227+
----------------------
228+
229+
Total msg sent - The total number of messages sent during the measurement
230+
period.
231+
232+
Total msg rcvd - The total number of messages received during the
233+
measurement period.
234+
235+
Total data sent - The total number of bytes sent during the measurement
236+
period.
237+
238+
Total data rcvd - The total number of bytes received during the measurement
239+
period.
240+
241+
Avg measure time - The average measurement time across all threads (µs).
242+
243+
Start variance - The maximum difference between the start times of all
244+
the threads (µs).
245+
246+
Run variance - The maximum difference between the measurement times of
247+
all the threads (µs).
248+
249+
Send throughput - The send throughput, aggregated across all connections,
250+
during the measurement period (bytes/second).
251+
252+
Recv throughput - The receive throughput, aggregated across all connections,
253+
during the measurement period (bytes/second).
254+
255+
Average throughput - The average of the send and receive throughputs.
256+
257+
In brief mode the output consists of a single line as follows:
258+
259+
info: results A,<nconn>,<sendthroughput>,<recvthroughput>,<proccpu>,<syscpu>
260+
261+
Network overheads
262+
-----------------
263+
264+
The message size that you specify defines the size of the 'application data'
265+
in each message sent or received. The actual amount of data for each message
266+
will be larger than this due to various network and protocol related overheads.
267+
268+
For IPv4/TCP, there is at least 28 bytes of overhead per message and in unusual
269+
cases this may be as much as 36 bytes. In addition the TCPv4 header is another
270+
24 bytes. In most cases, for this program, the IPv4 packet size will be <message
271+
size> + 52 bytes.
272+
273+
For IPv6/TCP, there is at least 72 bytes of overhead per message and in unusual
274+
cases this may be more due to additional header fields (each is 8 bytes). In
275+
most cases, for this program, the IPv6 packet size will be <message size> + 72
276+
bytes.
277+
278+
For Ethernet the standard Maximum Transmission Unit (MTU) is 1500 bytes. Each
279+
IP packet will be sent as a sequence of one or more Ethernet frames. Each frame
280+
has some overhead; normally this is 38 bytes but if 802.1q VLANs are being used
281+
it will be 42 bytes.
282+
283+
For WiFi the standard Maximum Transmission Unit (MTU) is 1500 bytes. Each IP
284+
packet will be sent as a sequence of one or more WiFi frames. Each frame has
285+
some overhead; normally this is 36 bytes but it may be as much as 44 bytes.
286+
287+
To give this some context, with the default test message size of 1024 bytes the
288+
IPv4 packet size will be 1076 bytes and the IPv6 packet size will be 1096 bytes.
289+
Both will therefore fit into a single Ethernet/WiFi frame. The associated
290+
Ethernet frame will be 1114 bytes for IPv4 and 1134 bytes for IPv6.
291+
292+
With a 1 Gbit/s Ethernet network, the maximum theoretical throughput will be
293+
112,208 frames/s for IPv4 and 110,229 frames/s for IPv6. This translates to
294+
a theoretical maximum application data rate of 109.5 Mbyte/s for IPv4 and
295+
107.6 Mbyte/s for IPv6.
296+

0 commit comments

Comments
 (0)