Testing the GreenCloud SoftBank and IIJ plans bought during the anniversary sale. ac01

Evening-peak mainland China speed tests are in section 4, and international connectivity is in section 5.

Contents

1. Configuration
2. Node Quality
3. itdog test
4. Evening peak speed test
5. International connectivity
6. Summary

1. Configuration

Plan 🇯🇵 GreenCloud Softbank 2222 🇯🇵 GreenCloud IIJ 2222
Location Japan, Tokyo Japan, Tokyo
CPU 2x EPYC 7763 vCPU 2x EPYC 7642 vCPU
RAM 4GB 4GB
Disk 20GB 20GB
Traffic 750GB(in+out) 750GB(in+out)
Bandwidth 10Gbps 10Gbps
Network 1 x IPv4,1 x /64 IPv6 1 x IPv4,1 x /64 IPv6
Price annually USD$22.00 annually USD$22.00

Looking Glass: https://greencloudvps.com/data-centers.php#

🔗 Quick links


YABS Node Quality nws.sh
[Softbank] | [IIJ] [Softbank] | [IIJ] [Softbank] | [IIJ]

2. Node Quality

bash <(curl -sL https://run.NodeQuality.com)

💻Basic information

Softbank IIJ

🎬IP quality

Softbank IIJ

🌐Network quality

Softbank IIJ

📍Return route

Softbank IIJ

3. TCP Ping(itdog.cn,2025-11-20 23:30)

Both have packet loss and jitter on all three networks. SoftBank has the worst Telecom outbound packet loss and jitter, while IIJ has the worst Unicom outbound packet loss and jitter.

Softbank IIJ

4. Evening peak proxy speed tests

During evening peak, Telecom on IIJ is stronger than SoftBank; SoftBank Telecom almost drops out. But this is very region-dependent and still far behind optimized routes.
Unicom on SoftBank is okay and has usable speed, while IIJ on Unicom is unusable.
Mobile depends on region. SoftBank can run very fast in some places and almost drops out in others, while IIJ generally cannot get much speed.

🇯🇵 Softbank


Node Average speed Maximum speed TLS RTT
ShanghaiTelecom@1Gbps 429.40Kbps 851.84Kbps 330ms
WenzhouTelecom@1Gbps 1.84Mbps 2.84Mbps 279ms
ShenzhenTelecom@1Gbps 450.50Kbps 524.29Kbps 269ms
GuangdongTelecom@1Gbps 1.19Mbps 1.83Mbps 139ms
GuangxiTelecom@1Gbps 101.19Kbps 196.61Kbps 243ms
Xi'anTelecom@1Gbps 9.07Mbps 17.08Mbps 445ms
ShanghaiUnicom@1Gbps 162.61Mbps 472.42Mbps 130ms
ShaoxingUnicom@1Gbps 63.49Mbps 169.11Mbps 92ms
ShandongUnicom@1Gbps 514.08Mbps 890.57Mbps 276ms
HebeiUnicom@1Gbps 90.68Mbps 163.56Mbps 417ms
GuangdongUnicom@1Gbps 34.58Mbps 123.64Mbps 272ms
HunanUnicom@1Gbps 77.14Mbps 89.60Mbps 250ms
ShanghaiMobile@1Gbps 254.91Mbps 340.99Mbps 100ms
LishuiMobile@1Gbps 265.37Mbps 385.66Mbps 82ms
GuangzhouMobile@1Gbps 507.83Kbps 655.25Kbps 198ms
GuangxiMobile@1Gbps 100.90Mbps 187.06Mbps 128ms
SichuanMobile@1Gbps 29.50Mbps 40.17Mbps 110ms
HebeiMobile@1Gbps 616.58Mbps 817.23Mbps 116ms

[Telecom speed-test records] | [Unicom speed-test records] | [Mobile speed-test records]

🇯🇵 IIJ


Node Average speed Maximum speed TLS RTT
ShanghaiTelecom@1Gbps 1.58Mbps 2.10Mbps 121ms
WenzhouTelecom@1Gbps 277.81Mbps 303.87Mbps 78ms
ShenzhenTelecom@1Gbps 51.23Mbps 72.29Mbps 128ms
GuangdongTelecom@1Gbps 3.65Mbps 5.36Mbps 88ms
GuangxiTelecom@1Gbps 34.42Mbps 53.23Mbps 124ms
Xi'anTelecom@1Gbps 131.74Mbps 296.94Mbps 171ms
ShanghaiUnicom@1Gbps 8.70Mbps 15.04Mbps 153ms
ShaoxingUnicom@1Gbps 6.37Mbps 11.44Mbps 98ms
ShandongUnicom@1Gbps 41.51Mbps 65.79Mbps 293ms
HebeiUnicom@1Gbps 3.21Mbps 4.40Mbps 330ms
GuangdongUnicom@1Gbps 768.68Kbps 1.18Mbps 131ms
HunanUnicom@1Gbps 319.45Kbps 393.22Kbps 446ms
ShanghaiMobile@1Gbps 760.50Kbps 1.05Mbps 152ms
LishuiMobile@1Gbps 6.58Mbps 14.32Mbps 316ms
GuangzhouMobile@1Gbps 835.46Kbps 982.78Kbps 238ms
GuangxiMobile@1Gbps 16.35Mbps 26.15Mbps 163ms
SichuanMobile@1Gbps 19.73Mbps 31.51Mbps 150ms
HebeiMobile@1Gbps 51.14Mbps 119.96Mbps 189ms

[Telecom speed-test records] | [Unicom speed-test records] | [Mobile speed-test records]

5. International connectivity - iPerf3 TCP IPv4 Single-threadspeed test (10 seconds per run, 2025-11-22 20:40-21:40)

Among local Tokyo nodes, SoftBank is generally much faster than IIJ, and speeds to Asia and the US West Coast are also clearly faster. Overall, SoftBank has better international connectivity.

🇯🇵 Softbank


Region Node Average/max upload (Mbps) Average/max download (Mbps) Latency (ms)
🇯🇵 Tokyo DMIT T1 3922 / 4080 3770 / 3958 1.72
🇯🇵 Tokyo BAGE JP 3751 / 3854 2424 / 3172 1.01
🇯🇵 Tokyo BAGE JPS 4278 / 4437 5139 / 5228 0.74
🇯🇵 Tokyo Nube 5570 / 5970 4746 / 5218 2.11
🇯🇵 Tokyo GreenCloud IIJ 2804 / 4728 6540 / 7186 0.68
🇯🇵 Tokyo DDPS CDN77 7859 / 8710 6771 / 7933 0.92
🇯🇵 Tokyo RFCHOST T1 5972 / 6481 7233 / 7696 0.86
🇯🇵 Tokyo YxVM VOL 8015 / 8624 7317 / 8295 1.05
🇭🇰 Hong Kong DMIT T1 2306 / 2549 2579 / 2811 48.85
🇭🇰 Hong Kong RFCHOST JINX 1023 / 1927 2952 / 3252 54.45
🇭🇰 Hong Kong RFCHOST CTC 3376 / 3679 3023 / 3359 48.35
🇭🇰 Hong Kong Nube 1891 / 2280 2393 / 2767 49.50
🇭🇰 Hong Kong GreenCloud DC1 3755 / 3966 3561 / 3981 44.75
🇭🇰 Hong Kong GreenCloud DC2 3199 / 3559 2633 / 2809 53.79
🇭🇰 Hong Kong YxVM VOL 3499 / 3740 2451 / 2868 47.55
🇭🇰 Hong Kong Hytron 1098 / 1208 2912 / 3343 47.99
🇭🇰 Hong Kong Halo STD 2437 / 2637 2493 / 2685 50.77
🇭🇰 Hong Kong Halo Lite 3315 / 3546 2833 / 3354 50.71
🇭🇰 Hong Kong BAGE HKS 973 / 1014 903 / 1003 50.84
🇭🇰 Hong Kong BAGE HKL 968 / 1001 873 / 962 47.49
🇭🇰 Hong Kong ClawCloud 987 / 1087 1018 / 1117 50.25
🇭🇰 Hong Kong Servps 816 / 1001 1052 / 1625 50.93
🇭🇰 Hong Kong Boil HKT 762 / 919 871 / 985 52.06
🇹🇼 Taiwan Misaka 2600 / 3077 2462 / 2768 46.92
🇸🇬 Singapore RFCHOST T1 1362 / 1551 1411 / 1701 82.04
🇸🇬 Singapore BAGE SG 2183 / 2674 1804 / 2018 67.84
🇸🇬 Singapore BAGE SGS 2373 / 2650 2247 / 2697 67.62
🇸🇬 Singapore GreenCloud DC1 2244 / 2555 1792 / 1976 70.00
🇸🇬 Singapore GreenCloud DC2 2338 / 2550 2083 / 2367 76.11
🇸🇬 Singapore IDC.Care 874 / 1006 851 / 1045 68.14
🇻🇳 Hanoi GreenCloud 1659 / 1872 1197 / 1395 85.90
🇮🇳 Mumbai HostDZire 357 / 487 276 / 389 172.63
🇦🇺 Sydney GreenCloud 1425 / 1773 1243 / 1379 99.36
🇺🇸 Los Angeles DMIT T1 532 / 686 355 / 482 110.78
🇺🇸 Los Angeles GreenCloud LA 1479 / 1771 1145 / 1370 99.69
🇺🇸 Los Angeles BAGE LA1 881 / 1006 689 / 830 99.80
🇺🇸 Los Angeles BAGE LA2 831 / 1007 644 / 911 108.73
🇺🇸 Los Angeles RFCHOST T1 384 / 723 540 / 731 109.37
🇺🇸 San Jose Nube 1475 / 1671 1163 / 1564 109.54
🇺🇸 San Jose HostDZire 1503 / 1721 1168 / 1315 104.95
🇺🇸 Seattle Misaka 1953 / 2201 1831 / 2140 83.27
🇺🇸 Salt Lake City BAGE 893 / 1049 611 / 749 117.73
🇷🇺 Moscow Misaka 528 / 745 395 / 533 248.92
🇩🇪 Frankfurt BAGE 562 / 814 433 / 570 250.36
🇩🇪 Frankfurt Nube 560 / 784 492 / 731 241.33
🇩🇪 Frankfurt BERO 603 / 821 564 / 781 228.14
🇩🇪 Nuremberg Netcup 477 / 727 465 / 713 262.99
🇬🇧 Coventry BAGE 244 / 413 290 / 504 262.94

[Test records]

🇯🇵 IIJ


Region Node Average/max upload (Mbps) Average/max download (Mbps) Latency (ms)
🇯🇵 Tokyo DMIT T1 3756 / 3967 2646 / 3003 2.84
🇯🇵 Tokyo BAGE JP 3293 / 3758 2269 / 2854 1.42
🇯🇵 Tokyo BAGE JPS 4073 / 4382 3437 / 4107 1.13
🇯🇵 Tokyo Nube 3982 / 4542 3248 / 3927 2.52
🇯🇵 Tokyo GreenCloud SB 6319 / 7257 3338 / 4716 1.22
🇯🇵 Tokyo DDPS CDN77 7472 / 8478 3549 / 4293 1.24
🇯🇵 Tokyo RFCHOST T1 5539 / 5956 3260 / 5228 1.15
🇯🇵 Tokyo YxVM VOL 7089 / 7999 3862 / 4919 1.34
🇭🇰 Hong Kong DMIT T1 2282 / 3038 1419 / 2211 53.15
🇭🇰 Hong Kong RFCHOST JINX 1061 / 2034 1858 / 2377 54.39
🇭🇰 Hong Kong RFCHOST CTC 2832 / 3225 1281 / 1849 53.39
🇭🇰 Hong Kong Nube 1357 / 1827 1261 / 1861 50.30
🇭🇰 Hong Kong GreenCloud DC1 3629 / 3996 1692 / 2536 44.99
🇭🇰 Hong Kong GreenCloud DC2 2466 / 2773 1291 / 2590 49.42
🇭🇰 Hong Kong YxVM VOL 2510 / 3319 1691 / 2127 45.70
🇭🇰 Hong Kong Hytron 1001 / 1210 1873 / 2300 51.25
🇭🇰 Hong Kong Halo STD 3017 / 3383 1727 / 2473 51.31
🇭🇰 Hong Kong Halo Lite 2687 / 3110 1337 / 2235 51.18
🇭🇰 Hong Kong BAGE HKS 958 / 1006 840 / 973 51.17
🇭🇰 Hong Kong BAGE HKL 959 / 1015 861 / 1034 49.60
🇭🇰 Hong Kong ClawCloud 943 / 1068 899 / 1166 59.82
🇭🇰 Hong Kong Servps 899 / 1007 1031 / 1535 46.27
🇭🇰 Hong Kong Boil HKT 744 / 870 785 / 940 51.90
🇹🇼 Taiwan Misaka 299 / 329 1935 / 2350 31.87
🇸🇬 Singapore RFCHOST T1 2187 / 2501 1559 / 2271 70.39
🇸🇬 Singapore BAGE SG 1781 / 2186 906 / 1712 67.88
🇸🇬 Singapore BAGE SGS 2193 / 2648 1795 / 2319 67.50
🇸🇬 Singapore GreenCloud DC1 1583 / 1990 1236 / 2283 67.86
🇸🇬 Singapore GreenCloud DC2 1868 / 2209 1467 / 2248 78.59
🇸🇬 Singapore IDC.Care 867 / 1035 440 / 904 68.82
🇻🇳 Hanoi GreenCloud 1384 / 1650 1675 / 2111 97.06
🇮🇳 Mumbai HostDZire 382 / 468 274 / 401 147.63
🇦🇺 Sydney GreenCloud 674 / 915 1254 / 1675 99.72
🇺🇸 Los Angeles DMIT T1 478 / 628 255 / 380 111.09
🇺🇸 Los Angeles GreenCloud LA 1428 / 1792 364 / 1783 98.98
🇺🇸 Los Angeles BAGE LA1 896 / 1007 598 / 756 99.10
🇺🇸 Los Angeles BAGE LA2 799 / 999 493 / 731 111.67
🇺🇸 Los Angeles RFCHOST T1 726 / 942 377 / 856 106.13
🇺🇸 San Jose Nube 1325 / 1685 979 / 1569 107.65
🇺🇸 San Jose HostDZire 1124 / 1386 1189 / 1733 103.10
🇺🇸 Seattle Misaka 1633 / 1859 1552 / 2007 88.79
🇺🇸 Salt Lake City BAGE 885 / 1012 326 / 483 125.27
🇷🇺 Moscow Misaka 480 / 720 235 / 496 265.02
🇩🇪 Frankfurt BAGE 550 / 745 443 / 654 233.14
🇩🇪 Frankfurt Nube 449 / 636 325 / 757 239.53
🇩🇪 Frankfurt BERO 387 / 544 378 / 678 267.72
🇩🇪 Nuremberg Netcup 419 / 738 449 / 659 277.51
🇬🇧 Coventry BAGE 326 / 543 284 / 534 235.14

[Test records]

6. Summary

For pure international connectivity, GreenCloud SoftBank is much stronger than IIJ.

If you are buying for direct connection and must choose between these two, pick SoftBank for Mobile/Unicom and IIJ for Telecom. SoftBank on Unicom can basically run; Mobile is pleasant in some places, though Guangdong and Sichuan could not run in the tests. IIJ is basically only pleasant for Telecom in some regions, and in many places it is still throttled to near drop-out during evening peak.

I do not really recommend GreenCloud directly for beginners, because performance is highly region-dependent; someone else's good experience may not be yours. ac07

For Asia-Pacific direct connection, the safer choice may simply be to pay more. ac02

7. About the tests

The machines tested in this article, including all machines used for iPerf3 speed tests, were purchased at my own expense. Results may be time-sensitive.
Kernel parameters of the tested machines (/etc/sysctl.conf) are as follows

fs.file-max = 6815744
net.ipv4.tcp_no_metrics_save=1
net.ipv4.tcp_ecn=0
net.ipv4.tcp_frto=0
net.ipv4.tcp_mtu_probing=0
net.ipv4.tcp_rfc1337=0
net.ipv4.tcp_sack=1
net.ipv4.tcp_fack=1
net.ipv4.tcp_window_scaling=1
net.ipv4.tcp_adv_win_scale=1
net.ipv4.tcp_moderate_rcvbuf=1
net.core.rmem_max=33554432
net.core.wmem_max=33554432
net.ipv4.tcp_rmem=4096 87380 33554432
net.ipv4.tcp_wmem=4096 16384 33554432
net.ipv4.udp_rmem_min=8192
net.ipv4.udp_wmem_min=8192
net.ipv4.ip_forward=1
net.ipv4.conf.all.route_localnet=1
net.ipv4.conf.all.forwarding=1
net.ipv4.conf.default.forwarding=1
net.core.default_qdisc=fq
net.ipv4.tcp_congestion_control=bbr
net.ipv6.conf.all.forwarding=1
net.ipv6.conf.default.forwarding=1

8. Acknowledgements

Thanks to “Impart Speed Test“(@Kyoma) for providing public-interest speed-test service.