TOFFEE

DOCUMENTATION :: TEST CASES :: TEST RESULTS :: TOFFEE-Mocha-1.0.14 Development version

Written by: Kiran Kankipati
Published: 07-Jul-2016


Here are the TOFFEE-Mocha test cases and test results of the upcoming new TOFFEE-Mocha which is still under development. The features of this TOFFEE-Mocha are discussed in the software development update: TOFFEE-Mocha WAN Emulation software development - Update: 1-July-2016

Test case1 :: 999 millisecond constant packet delay: As you can see unlike 40 milliseconds the maximum limit which existed earlier, the new 999 milliseconds delay range allows users to slow down the transfer rates even further.

kiran@HP-ENVY-15:~/temp$ ping 192.168.0.1 -s 1000
PING 192.168.0.1 (192.168.0.1) 1000(1028) bytes of data.
1008 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=2000 ms
1008 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=2000 ms
1008 bytes from 192.168.0.1: icmp_seq=3 ttl=64 time=2000 ms
1008 bytes from 192.168.0.1: icmp_seq=4 ttl=64 time=2000 ms
1008 bytes from 192.168.0.1: icmp_seq=5 ttl=64 time=2998 ms
1008 bytes from 192.168.0.1: icmp_seq=6 ttl=64 time=2997 ms
1008 bytes from 192.168.0.1: icmp_seq=7 ttl=64 time=3995 ms
1008 bytes from 192.168.0.1: icmp_seq=8 ttl=64 time=3985 ms
1008 bytes from 192.168.0.1: icmp_seq=9 ttl=64 time=3984 ms
1008 bytes from 192.168.0.1: icmp_seq=10 ttl=64 time=3984 ms
1008 bytes from 192.168.0.1: icmp_seq=11 ttl=64 time=3983 ms
1008 bytes from 192.168.0.1: icmp_seq=12 ttl=64 time=3982 ms
1008 bytes from 192.168.0.1: icmp_seq=13 ttl=64 time=3984 ms
1008 bytes from 192.168.0.1: icmp_seq=14 ttl=64 time=3982 ms
^C
--- 192.168.0.1 ping statistics ---
18 packets transmitted, 14 received, 22% packet loss, time 17007ms
rtt min/avg/max/mdev = 2000.042/3277.214/3995.537/873.965 ms, pipe 4
kiran@HP-ENVY-15:~/temp$

Test case2 :: 500 millisecond constant packet delay: With 500 milliseconds you get roughly double the performance of 999 milliseconds.

kiran@HP-ENVY-15:~/temp$ ping 192.168.0.1 -s 1000
PING 192.168.0.1 (192.168.0.1) 1000(1028) bytes of data.
1008 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=3 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=4 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=5 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=6 ttl=64 time=1488 ms
1008 bytes from 192.168.0.1: icmp_seq=7 ttl=64 time=1481 ms
1008 bytes from 192.168.0.1: icmp_seq=8 ttl=64 time=1481 ms
1008 bytes from 192.168.0.1: icmp_seq=9 ttl=64 time=1008 ms
1008 bytes from 192.168.0.1: icmp_seq=10 ttl=64 time=1002 ms
^C
--- 192.168.0.1 ping statistics ---
11 packets transmitted, 10 received, 9% packet loss, time 10017ms
rtt min/avg/max/mdev = 1002.077/1147.151/1488.063/220.133 ms, pipe 2
kiran@HP-ENVY-15:~/temp$

Test case3 :: 500 millisecond constant packet delay + random packet delay: With constant delay (in this case 500 milliseconds) if you enable the new random packet delay feature, it will skip delay randomly few packets. Which can be controlled via random delay factor. In this case the random delay factor value is set to 1. And you can see below few packets are not delayed. Hence their ping response time almost reduced to half (i.e around 500 ms).

kiran@HP-ENVY-15:~/temp$ ping 192.168.0.1 -s 1000
PING 192.168.0.1 (192.168.0.1) 1000(1028) bytes of data.
1008 bytes from 192.168.0.1: icmp_seq=1 ttl=64 time=1503 ms
1008 bytes from 192.168.0.1: icmp_seq=2 ttl=64 time=1497 ms
1008 bytes from 192.168.0.1: icmp_seq=3 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=4 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=5 ttl=64 time=1001 ms
1008 bytes from 192.168.0.1: icmp_seq=6 ttl=64 time=1001 ms
1008 bytes from 192.168.0.1: icmp_seq=7 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=8 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=9 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=10 ttl=64 time=419 ms
1008 bytes from 192.168.0.1: icmp_seq=11 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=12 ttl=64 time=1001 ms
1008 bytes from 192.168.0.1: icmp_seq=13 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=14 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=15 ttl=64 time=1001 ms
1008 bytes from 192.168.0.1: icmp_seq=16 ttl=64 time=502 ms
1008 bytes from 192.168.0.1: icmp_seq=17 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=18 ttl=64 time=502 ms
1008 bytes from 192.168.0.1: icmp_seq=19 ttl=64 time=1002 ms
1008 bytes from 192.168.0.1: icmp_seq=20 ttl=64 time=1001 ms
1008 bytes from 192.168.0.1: icmp_seq=21 ttl=64 time=1002 ms
^C
--- 192.168.0.1 ping statistics ---
22 packets transmitted, 21 received, 4% packet loss, time 21029ms
rtt min/avg/max/mdev = 419.093/974.135/1503.026/250.662 ms, pipe 2
kiran@HP-ENVY-15:~/temp$

Random Packet delay: As discussed in my VLOG/update earlier, the idea of Random packet delay is to introduce the fluctuating, bursty nature of packet flow. So here are various tests done which shows the same in action. These tests below are performed while downloading a large file by enabling random packet delay along with various values of constant packet delay.

Test case4 :: 2 millisecond constant packet delay + random packet delay: With constant delay of 2 millisecond and random packet delay you can notice the blue curve which almost appears constant. The traffic in this case is bursty but it is not that significant to notice in the graph shown below.
TOFFEE_Mocha_2ms_delay_with_random_packet_delay

Test case5 :: 10 millisecond constant packet delay + random packet delay: With constant delay of 10 millisecond and random packet delay you can notice the blue curve which almost appears constant. The traffic in this case is bursty but it is not that significant to notice in the graph shown below. But it appears somewhat fluctuating than the 5 millisecond test case4 above.
TOFFEE_Mocha_10ms_delay_with_random_packet_delay

Test case6 :: 200 millisecond constant packet delay + random packet delay: With constant delay of 200 millisecond and random packet delay you can notice the fluctuating blue curve. With this we can understand the true purpose of random packet delay.
TOFFEE_Mocha_200ms_delay_with_random_packet_delay

Test case7 :: 200 millisecond constant packet delay + WITHOUT random packet delay: With constant delay of 200 millisecond and WITHOUT random packet delay feature enabled you can notice the steady blue curve. This is a direct comparison of a test case with constant packet delay 200 millisecond with and without random packet delay. With random packet delay it makes the network performance choppy, fluctuating and bursty, but without random packet delay feature the network performance appears almost constant.
TOFFEE_Mocha_200ms_delay_without_random_packet_delay

So in my next upcoming TOFFEE-Mocha release I may include all these new features and updated old features. If you are in need of any specific feature (or scenario) you can kindly let know. If plausible and feasible I can support the same and release as a part of my upcoming TOFFEE-Mocha release. Kindly stay tuned !


Thank you, Kiran




Back to Top