Benchmarking Network Performance

Tool Used: iperf

First Instance ECS on Cloud:

root@iZj6c5vmzbgujq3xi2m45aZ:~# phoronix-test-suite system-info

Phoronix Test Suite v5.2.1

System Information

Hardware:

Processor: Intel Xeon E5-2682 v4 @ 2.50GHz (1 Core), Motherboard: Alibaba Cloud ECS, Chipset: Intel 440FX- 82441FX PMC, Memory: 1 x 1024 MB RAM, Disk: 40GB, Graphics: Cirrus Logic GD 5446, Network: Red Hat Virtio device

Software:

OS: Ubuntu 18.04, Kernel: 4.15.0-45-generic (x86_64), File-System: ext4, Screen Resolution: 1024x768, System Layer: KVM

Second similar Instance on Azure Cloud:

root@testbenchmark:~# phoronix-test-suite system-info

Phoronix Test Suite v5.2.1

System Information

Hardware:

Processor: Intel Xeon Platinum 8171M @ 2.10GHz (1 Core), Motherboard: Microsoft Virtual Machine v7.0, Chipset: Intel 440BX/ZX/DX, Memory: 1 x 1024 MB Microsoft, Disk: 32GB Virtual Disk + 4GB Virtual Disk, Graphics: Microsoft Hyper-V virtual VGA

Software:

OS: Ubuntu 18.04, Kernel: 5.0.0-1035-azure (x86_64), File-System: ext4, Screen Resolution: 1152x864, System Layer: Microsoft Hyper-V Server

	ECS	Azure VM
Processor	Intel Xeon E5-2682 v4 @ 2.50GHz (1 Core)	Intel Xeon Platinum 8171M @ 2.10GHz (1 Core)
Motherboard	Alibaba Cloud ECS	Microsoft Virtual Machine v7.0
Chipset	Intel 440FX- 82441FX PMC	Intel 440BX/ZX/DX
Memory	1 x 1024 MB RAM	1 x 1024 MB Microsoft
Disk	40GB	32GB Virtual Disk + 4GB Virtual Disk
Graphics	Cirrus Logic GD 5446	Microsoft Hyper-V virtual VGA
Network	Red Hat Virtio device
OS	Ubuntu 18.04	Ubuntu 18.04
Kernel	4.15.0-45-generic (x86_64)	5.0.0-1035-azure (x86_64)
File-System	ext4	ext4
Screen Resolution	1024x768	1152x864
System Layer	KVM	Microsoft Hyper-V Server

NOTE: The test and comparison here is very high level and the purpose here is to show how I do usually benchmarking and reporting.

Network workload

To install iperf on ubuntu linux use ‘sudo apt-get install -y iperf’

iPerf tool requires two systems because one system must act as a server, while the other acts as a client. The client connects to the server we are testing the speed of.

Sender Server:

ECS on cloud: 47.75.190.46 (Internet IP)

Client Listening Server:

VM on Azure Cloud : 138.91.242.171 (Public IP) à Allowed UDP/TCP communication over 5001

root@testwork:~# iperf -s

------------------------------------------------------------

Server listening on TCP port 5001

TCP window size: 128 KByte (default)

------------------------------------------------------------

[ 4] local 10.0.0.4 port 5001 connected with 47.75.190.46 port 50792

[ ID] Interval Transfer Bandwidth

[ 4] 0.0-16.6 sec 2.25 MBytes 1.14 Mbits/sec

root@iZj6c5vmzbgujq3xi2m45aZ:~# iperf -c 138.91.242.171

------------------------------------------------------------

Client connecting to 138.91.242.171, TCP port 5001

TCP window size: 85.0 KByte (default)

------------------------------------------------------------

[ 3] local 172.31.146.175 port 50792 connected with 138.91.242.171 port 5001

[ ID] Interval Transfer Bandwidth

[ 3] 0.0-10.2 sec 2.25 MBytes 1.85 Mbits/sec

From the above output, we can see that i got a speed of 1.85Mbits/sec. The output shows something more.

Interval: Interval specifies the time duration for which the data is transferred.

Transfer: This column shows the transferred data size.

Bandwidth: This shows the rate of speed with which the data is transferred.

Using iPerf, we can also test the maximum throughput achieved via UDP connections.

root@testwork:~# iperf -s -u

------------------------------------------------------------

Server listening on UDP port 5001

Receiving 1470 byte datagrams

UDP buffer size: 208 KByte (default)

------------------------------------------------------------

[ 3] local 10.0.0.4 port 5001 connected with 47.75.190.46 port 39304

[ ID] Interval Transfer Bandwidth Jitter Lost/Total Datagrams

[ 3] 0.0-10.0 sec 1.25 MBytes 1.05 Mbits/sec 0.305 ms 0/ 893 (0%)

root@iZj6c5vmzbgujq3xi2m45aZ:~# iperf -c 138.91.242.171 -u

------------------------------------------------------------

Client connecting to 138.91.242.171, UDP port 5001

Sending 1470 byte datagrams, IPG target: 11215.21 us (kalman adjust)

UDP buffer size: 208 KByte (default)

------------------------------------------------------------

[ 3] local 172.31.146.175 port 39304 connected with 138.91.242.171 port 5001

[ ID] Interval Transfer Bandwidth

[ 3] 0.0-10.0 sec 1.25 MBytes 1.05 Mbits/sec

[ 3] Sent 893 datagrams

[ 3] Server Report:

[ 3] 0.0-10.0 sec 1.25 MBytes 1.05 Mbits/sec 0.000 ms 0/ 893 (0%)

iPerf limits the bandwidth for UDP clients to 1 Mbit per second by default. We can change this with the -b flag, replacing the number after with the maximum bandwidth rate we wish to test against. If we are testing for network speed, set this number above the maximum bandwidth cap provided by receiver server:

iperf -c 138.91.242.171 -u -b 1000m

This tells the client that we want to achieve a maximum of 1000 Mbits per second if possible.

root@testwork:~# iperf -s -u -b 1000m

------------------------------------------------------------

Server listening on UDP port 5001

Receiving 1470 byte datagrams

UDP buffer size: 208 KByte (default)

------------------------------------------------------------

[ 3] local 10.0.0.4 port 5001 connected with 47.75.190.46 port 47812

[ ID] Interval Transfer Bandwidth Jitter Lost/Total Datagrams

[ 3] 0.0-10.1 sec 1.43 MBytes 1.19 Mbits/sec 3.330 ms 662244/663261 (1e+02%)

root@iZj6c5vmzbgujq3xi2m45aZ:~# iperf -c 138.91.242.171 -u -b 1000m

------------------------------------------------------------

Client connecting to 138.91.242.171, UDP port 5001

Sending 1470 byte datagrams, IPG target: 11.76 us (kalman adjust)

UDP buffer size: 208 KByte (default)

------------------------------------------------------------

[ 3] local 172.31.146.175 port 47812 connected with 138.91.242.171 port 5001

[ ID] Interval Transfer Bandwidth

[ 3] 0.0-10.0 sec 930 MBytes 777 Mbits/sec

[ 3] Sent 663261 datagrams

[ 3] Server Report:

[ 3] 0.0-10.1 sec 1.43 MBytes 1.19 Mbits/sec 0.000 ms 662244/663261 (0%)

We may also want to test both servers for the maximum amount of throughput. This can easily be done using the built-in bidirectional testing feature iPerf offers. Also we can use below parameter to get maximum network load, this will sum the overall result at end.

· -w - increase the TCP window size;

· -t 30 – is the time in seconds for the test to be done (by default, it is 10 seconds)

· -P 8 – is the number of parallel threads (streams) to get the maximum channel load

Network Outbound Bandwidth: