Changes between Version 23 and Version 24 of PerformanceTesting


Ignore:
Timestamp:
11/11/11 09:44:40 (12 years ago)
Author:
Samuli Seppänen
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • PerformanceTesting

    v23 v24  
    117117All EC2 instances were created to the same availability zone (us-east-1d). The setup of 100 client instances took ~4 minutes with 34 threads running Fabric (~3 runs per thread).
    118118
    119 == OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/5 clients (MP-1) ==
     119== MP-1 results ==
     120
     121=== OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/5 clients (MP-1) ===
    120122
    121123'''Test name:''' 07-mp1-m1.small-openvpn221
     
    143145OpenVPN 2.2.1 on an Amazon EC2 m1.small instance can handle 5 OpenVPN client instances well.
    144146
    145 == OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/20 clients (MP-2) ==
     147== MP-2 results ==
     148
     149=== OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/20 clients (MP-2) ===
    146150
    147151'''Test name:'''  13-mp2-m1.small-openvpn221
     
    172176OpenVPN 2.2.1 on an Amazon EC2 m1.small instance seems to be able to handle 20 OpenVPN clients fairly well.
    173177
    174 == OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/50 clients (MP-3) ==
     178=== OpenVPN 2.2.1/m1.large/UDP/IPv4/OpenSSL/20 clients (MP-2) ===
     179
     180'''Test name:'''  16-mp2-m1.large-openvpn221
     181
     182'''Connection establishment'''
     183
     184Client initialization phase took 2 seconds, judging from server's CPU load and network traffic levels. CPU utilization was 1.5-3.5% during this time. 
     185
     186'''Client statistics'''
     187
     188||'''Test'''||'''Subtest'''||'''Host'''||'''Clients'''||'''Total transfer (MB)'''||'''Total bandwidth (MB/s)'''||'''Average bandwidth (MB/s)'''||
     189||16-mp2-m1.large-openvpn221||10s||average||19||386.42||37.06||1.95053||
     190||16-mp2-m1.large-openvpn221||30s||average||20||339.85||9.27||0.4635||
     191||16-mp2-m1.large-openvpn221||60s||average||20||1168.61||18.91||0.9455||
     192
     193'''Server statistics'''
     194
     195||'''Test'''||'''Subtest'''||'''Real length'''||'''CPU usr (%)'''||'''CPU sys (%)'''||'''CPU total (%)'''||'''CPU wait (%)'''||
     196||16-mp2-m1.large-openvpn221||10s||22||8.34841||14.974||32.1606||0.0163636||
     197||16-mp2-m1.large-openvpn221||30s||72||8.82207||15.8551||30.9363||0.00511111||
     198||16-mp2-m1.large-openvpn221||60s||72||8.73214||15.6736||31.3002||0.252833||
     199
     200'''Analysis'''
     201
     202A ''m1.large'' instance can easily handle 20 clients. As tests were more precisely timed (with "at") than in earlier MP-2 tests (which used chained commands with "sleep" in between), the overly long test periods must have been caused by the iperf server not being able to handle client connections quickly enough.
     203
     204== MP-3 results ==
     205
     206=== OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/50 clients (MP-3) ===
    175207
    176208'''Test name:'''  14-mp3-m1.small-openvpn221
     
    197229Same as for 100 client version of this test, except that the fake performance increase in the last test is not as pronounced.
    198230
    199 == OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/100 clients (MP-4) ==
     231== MP-4 results ==
     232
     233=== OpenVPN 2.2.1/m1.small/UDP/IPv4/OpenSSL/100 clients (MP-4) ===
    200234
    201235'''Test name:'''  15-mp4-m1.small-openvpn221
     
    294328[http://iperf.sourceforge.net/ Iperf] can be used to generate network traffic. It has previously been used to tune OpenVPN performance: look [wiki:Gigabit_Networks_Linux here].
    295329
     330'''NOTE:''' if multiple clients are connecting simultaneously to an iperf server, the delay in connection establishment can be considerable. For example, a 30 second iperf burst with 20 clients can take 72 seconds (for details look [wiki:PerformanceTesting#MP-2results here]).
     331
    296332== Collecting data ==
    297333
     
    299335
    300336
    301