V2ray 不完全测速

背景

最近看到 v2fly 上面多了蛮多新玩法。综合个人隐蔽性需求，将对以下搭配使用 iperf, speedtest mini, v2ray, clash premium 和 speedtest-cli 进行测速。

测速方式来自：https://steemit.com/cn/@v2ray/3cjiux

只考虑这两种隐蔽性强和兼容相对广泛的搭配。

比如 VLESS+WEB 通过分流器的操作，识别起来比较方便，WireShark就可以很简单的区分，故不考虑。

HTTP 协议	TLS	v2ray inbound协议	加密
HTTP/2	YES	H2C	none
HTTP/2	YES	H2C	AES-128-GCM
HTTP 1.x	YES	WS	none
HTTP 1.x	YES	WS	AES-128-GCM

本次测试的目的

测试 VMess 基于 HTTP 隐藏的各种加密方式，在常规传输过程中的参考速度。

测试方法

在两台 2c4h16g 4Gbps 的 Oracle VM.Standard.E4.Flex 虚拟机上面分别搭建基于 HTTP 1.x 和 HTTP/2 v2ray 服务端以及 speedtest mini服务端，并在一台2c4h16g 4Gbps 的 Oracle VM.Standard3.Flex 虚拟机搭建服务端，使用 speedtest-cli 通过代理进行测速。
如何判断真的走了代理？查看服务端和客户端的网卡流量信息和 v2ray 日志信息

测试环境

V2Ray Core 版本 v4.44.0

Clash 版本：Clash 2022.01.27 linux amd64 with go1.18beta1 Thu Jan 27 13:02:53 UTC 2022

speedtest-cli 2.1.2 Python 3.8.10 (default, Nov 26 2021, 20:14:08) [GCC 9.3.0]

系统均使用 Ubuntu 20.04，并开启 BBR

VM 都设置为 CPU 可突发，突发基线为 50%

2 x Oracle VM.Standard.E4.Flex 2c4h16g 4Gbps

Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   40 bits physical, 48 bits virtual
CPU(s):                          4
On-line CPU(s) list:             0-3
Thread(s) per core:              2
Core(s) per socket:              2
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       AuthenticAMD
CPU family:                      25
Model:                           1
Model name:                      AMD EPYC 7J13 64-Core Processor
Stepping:                        1
CPU MHz:                         2445.406
BogoMIPS:                        4890.81
Virtualization:                  AMD-V
Hypervisor vendor:               KVM
Virtualization type:             full
L1d cache:                       128 KiB
L1i cache:                       128 KiB
L2 cache:                        1 MiB
L3 cache:                        16 MiB
NUMA node0 CPU(s):               0-3
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Full AMD retpoline, IBPB conditional, IBRS_FW, STIBP conditional, RSB filli
                                 ng
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

1 x Oracle VM.Standard3.Flex 2c4h16g 4Gbps

Architecture:                    x86_64
CPU op-mode(s):                  32-bit, 64-bit
Byte Order:                      Little Endian
Address sizes:                   40 bits physical, 57 bits virtual
CPU(s):                          4
On-line CPU(s) list:             0-3
Thread(s) per core:              2
Core(s) per socket:              2
Socket(s):                       1
NUMA node(s):                    1
Vendor ID:                       GenuineIntel
CPU family:                      6
Model:                           106
Model name:                      Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz
Stepping:                        6
CPU MHz:                         2593.966
BogoMIPS:                        5187.93
Virtualization:                  VT-x
Hypervisor vendor:               KVM
Virtualization type:             full
L1d cache:                       64 KiB
L1i cache:                       64 KiB
L2 cache:                        8 MiB
L3 cache:                        16 MiB
NUMA node0 CPU(s):               0-3
Vulnerability Itlb multihit:     Not affected
Vulnerability L1tf:              Not affected
Vulnerability Mds:               Not affected
Vulnerability Meltdown:          Not affected
Vulnerability Spec store bypass: Mitigation; Speculative Store Bypass disabled via prctl and seccomp
Vulnerability Spectre v1:        Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:        Mitigation; Enhanced IBRS, IBPB conditional, RSB filling
Vulnerability Srbds:             Not affected
Vulnerability Tsx async abort:   Not affected

测速工具

使用 iperf3 进行服务器之间裸连测速
使用 speedtest-cli 进行代理测速
测速节点搭建见 → https://iitii.github.io/2022/02/27/1/

基本信息

Host	v2ray	speedtest-mini
n3kr.google.com	server	server
n4kr.google.com	server	server
n5kr.google.com	client	client

iperf3 可以查看客户端和服务端为 4Gbps 的带宽
speedtest-cli 的测速结果也基本如此

序号	上传	下载
0	3695.28	2651.07
1	3743	2528.95
2	3737.46	2526.12
3	3720.5	2518.34
4	3707.66	2169.3
5	3696.48	2668.12
6	3740.51	2534.96
7	3623.87	2178.24
8	3651.62	2278.35
9	3681.29	2627.53

Vmess 测速

测速命令

three (){
for i  in {1..$1}
do 
 eval $2
 sleep 5
done
}
three 10 "speedtest-cli --mini http://n3kr.google.com --csv"
# speedtest-cli --mini http://138.2.114.253:80 --csv-header

测速节点使用 speedtest mini 搭建，详见 →

h2 测速的过程中原版客户端出了点奇奇怪怪的问题，后续都换成了 clash premium 进行测试

HTTP/2+TLS+H2C+AUTO+WEB

测试过程中此种测试模式出现过服务端被打挂的情况

https://github.com/Dreamacro/clash/releases/tag/premium

three 20 "speedtest-cli --mini http://n4kr.google.com --csv"

序号	下载	上传
0	1076.26	2231.14
1	1078.07	1844.21
2	1065.16	2322.42
3	1091.96	2508.72
4	1066.28	1718.2
5	1087.6	1513
6	1047.8	1628.8
7	1091.55	1880.86
8	1076.29	1597.85
9	1088.29	2505.67
10	1080.98	2045.66
11	1061.3	2531.48
12	1071.67	1211.68
13	1061.07	2367.3
14	1080.41	1783.14
15	1068.84	1295.37
16	1084.67	1853.48
17	1070.99	2470.81
18	1056.12	1826.57
19	1075.17	1855.11

HTTP/2+TLS+H2C+NONE+WEB