Testing the GreenCloud SG DC1 and DC2 plans bought during the anniversary sale. 
I previously tested GreenCloud Japan SoftBank and IIJ; see:
Comparing GreenCloud SoftBank / IIJ | Tokyo, Japan | Evening Peak Speed Tests | International Connectivity Review
Contents
1. Configuration
2. Node Quality
3. Streaming unlock
4. itdog test
5. Evening peak speed test
6. International connectivity
7. Summary
1. Configuration
| Plan | 🇸🇬 GreenCloud DC1 2222 | 🇸🇬 GreenCloud DC2 2222 |
|---|---|---|
| Location | Singapore | Singapore |
| CPU | 2 x AMD EPYC Milan vCPU | 2 x AMD Ryzen 3960X vCPU |
| RAM | 2GB | 2GB |
| Disk | 22GB | 22GB |
| 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 | Streaming unlock |
|---|---|---|---|
| [DC1] | [DC2] | [DC1] | [DC2] | [DC1] | [DC2] | [DC1] | [DC2] |
DC1 upstream is xTom; DC2 upstream is Leaseweb.
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
2. Node Quality
bash <(curl -sL https://run.NodeQuality.com)
💻Basic information
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
🎬IP quality
Anycast IPs, but unlocking is currently quite good.
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
🌐Network quality
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
📍Return route
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
3. Streaming unlock
bash <(curl -L -s check.unlock.media)
| 🇸🇬 DC1 | 🇸🇬 DC2 |
|---|---|
4. TCP Ping(itdog.cn,2025-12-01 23:00)
Judging by this latency, they are basically not suitable for direct connection, especially DC2, where all three networks are orange.
🇸🇬 DC1
| China Telecom | China Unicom | China Mobile |
|---|---|---|
🇸🇬 DC2
| China Telecom | China Unicom | China Mobile |
|---|---|---|
5. Evening peak proxy speed tests (2025-12-01 23:00)
DC1 Mobile looks green, but cannot really run; DC2 cannot run on any of the three networks.
🇸🇬 DC1
| Node | Average speed | Maximum speed | TLS RTT |
|---|---|---|---|
| ShenzhenTelecom@1Gbps | 2.04Mbps | 4.82Mbps | 472ms |
| GuangdongTelecom@1Gbps | 506.56Kbps | 655.36Kbps | 470ms |
| ShanghaiTelecom@1Gbps | 396.67Kbps | 917.38Kbps | 491ms |
| WenzhouTelecom@1Gbps | 491.47Kbps | 851.90Kbps | 450ms |
| GuangdongUnicom@1Gbps | 995.79Kbps | 1.41Mbps | 482ms |
| ShanghaiUnicom@1Gbps | 39.71Mbps | 182.67Mbps | 457ms |
| ShaoxingUnicom@1Gbps | 1.49Mbps | 2.41Mbps | 1285ms |
| HebeiUnicom@1Gbps | 187.17Mbps | 385.53Mbps | 227ms |
| GuangzhouMobile@1Gbps | 957.33Kbps | 1.07Mbps | 245ms |
| GuangxiMobile@1Gbps | 246.63Mbps | 396.88Mbps | 180ms |
| ShanghaiMobile@1Gbps | 9.47Mbps | 17.57Mbps | 279ms |
| LishuiMobile@1Gbps | 36.07Mbps | 50.56Mbps | 93ms |
🇸🇬 DC2
| Node | Average speed | Maximum speed | TLS RTT |
|---|---|---|---|
| ShenzhenTelecom@1Gbps | 2.18Mbps | 6.88Mbps | 471ms |
| GuangdongTelecom@1Gbps | 54.37Kbps | 144.80Kbps | 534ms |
| ShanghaiTelecom@1Gbps | 810.30Kbps | 2.42Mbps | 332ms |
| WenzhouTelecom@1Gbps | 1.29Mbps | 3.79Mbps | 607ms |
| GuangdongUnicom@1Gbps | 2.34Mbps | 8.67Mbps | 502ms |
| ShanghaiUnicom@1Gbps | 256.72Mbps | 531.11Mbps | 377ms |
| ShaoxingUnicom@1Gbps | 1.07Mbps | 2.95Mbps | 389ms |
| HebeiUnicom@1Gbps | 180.98Kbps | 327.68Kbps | 372ms |
| GuangzhouMobile@1Gbps | 101.19Kbps | 327.68Kbps | 388ms |
| GuangxiMobile@1Gbps | 1.85Mbps | 5.32Mbps | 392ms |
| ShanghaiMobile@1Gbps | 2.26Mbps | 6.13Mbps | 486ms |
| LishuiMobile@1Gbps | 1.36Mbps | 5.08Mbps | 391ms |
[Telecom speed-test records] | [Unicom speed-test records] | [Mobile speed-test records]
6. International connectivity - iPerf3 TCP IPv4 Single-threadspeed test (10 seconds per run, 2025-12-02 01:00-02:00)
DC1 local speed is higher than DC2. To Hong Kong, DC1 is also better than DC2 with lower latency. DC2 to Japan CTC routes are much worse than DC1, though probably nobody pulls Singapore from Japan.
🇸🇬 DC1
| Region | Node | Average/max upload (Mbps) | Average/max download (Mbps) | Latency (ms) |
|---|---|---|---|---|
| 🇸🇬 Singapore | RFCHOST T1 | 8410 / 8549 | 8462 / 8599 | 0.78 |
| 🇸🇬 Singapore | BAGE SG | 3627 / 3842 | 2911 / 5042 | 0.82 |
| 🇸🇬 Singapore | BAGE SGS | 4395 / 4557 | 4317 / 5212 | 1.18 |
| 🇸🇬 Singapore | Nube | 3498 / 4439 | 5012 / 5281 | 1.62 |
| 🇸🇬 Singapore | IDC.Care | 904 / 992 | 901 / 1050 | 1.27 |
| 🇭🇰 Hong Kong | DMIT T1 | 2986 / 3320 | 2829 / 2960 | 30.46 |
| 🇭🇰 Hong Kong | RFCHOST JINX | 1387 / 2004 | 3854 / 4075 | 30.78 |
| 🇭🇰 Hong Kong | RFCHOST CTC | 4358 / 4564 | 4294 / 4516 | 30.54 |
| 🇭🇰 Hong Kong | Nube | 1762 / 1976 | 4060 / 4449 | 30.52 |
| 🇭🇰 Hong Kong | GreenCloud DC1 | 4646 / 4894 | 4401 / 4862 | 36.61 |
| 🇭🇰 Hong Kong | GreenCloud DC2 | 5071 / 5705 | 3762 / 4239 | 30.29 |
| 🇭🇰 Hong Kong | HostHatch | 5460 / 5749 | 4149 / 4457 | 30.75 |
| 🇭🇰 Hong Kong | YxVM VOL | 5472 / 6124 | 4237 / 4472 | 30.69 |
| 🇭🇰 Hong Kong | FxTransit | 5021 / 5807 | 4066 / 4433 | 31.77 |
| 🇭🇰 Hong Kong | Hytron | 587 / 1263 | 3684 / 4171 | 40.84 |
| 🇭🇰 Hong Kong | Halo STD | 4745 / 4982 | 4324 / 4510 | 30.36 |
| 🇭🇰 Hong Kong | Halo Lite | 4722 / 4967 | 3544 / 4120 | 30.38 |
| 🇭🇰 Hong Kong | BAGE HKS | 994 / 1009 | 1011 / 1102 | 30.36 |
| 🇭🇰 Hong Kong | BAGE HKL | 987 / 1000 | 967 / 1077 | 30.43 |
| 🇭🇰 Hong Kong | ClawCloud | 920 / 1066 | 1033 / 1052 | 40.43 |
| 🇭🇰 Hong Kong | Servps | 935 / 992 | 1071 / 1630 | 34.90 |
| 🇭🇰 Hong Kong | Boil HKT | 897 / 921 | 910 / 935 | 31.72 |
| 🇭🇰 Hong Kong | Kaze HKBN | 1891 / 2051 | 2022 / 2183 | 31.37 |
| 🇹🇼 Taipei | Misaka | 1607 / 1921 | 1869 / 2223 | 74.03 |
| 🇹🇼 Taichung | HiNet | 475 / 699 | 712 / 778 | 85.73 |
| 🇯🇵 Tokyo | DMIT T1 | 2405 / 2636 | 1848 / 2027 | 67.19 |
| 🇯🇵 Tokyo | BAGE JP | 2293 / 2623 | 1862 / 2351 | 66.73 |
| 🇯🇵 Tokyo | BAGE JPS | 2462 / 2730 | 1858 / 2038 | 67.21 |
| 🇯🇵 Tokyo | Nube | 1893 / 2067 | 2371 / 2623 | 68.31 |
| 🇯🇵 Tokyo | GreenCloud SB | 1528 / 1971 | 2245 / 2559 | 71.22 |
| 🇯🇵 Tokyo | GreenCloud IIJ | 1138 / 2131 | 1299 / 2176 | 67.79 |
| 🇯🇵 Tokyo | RFCHOST T1 | 2361 / 2619 | 1789 / 1963 | 73.82 |
| 🇯🇵 Tokyo | YxVM VOL | 2445 / 2723 | 1902 / 2042 | 66.87 |
| 🇯🇵 Tokyo | FxTransit | 128 / 160 | 1689 / 1833 | 74.76 |
| 🇰🇷 Seoul | AWS Lightsail | 703 / 1368 | 1821 / 1997 | 85.16 |
| 🇻🇳 Hanoi | GreenCloud | 3957 / 4249 | 4005 / 4215 | 42.39 |
| 🇻🇳 Ho Chi Minh City | GreenCloud | 3178 / 3919 | 3412 / 3910 | 38.42 |
| 🇮🇳 Mumbai | HostDZire | 431 / 545 | 398 / 490 | 67.90 |
| 🇦🇺 Sydney | GreenCloud | 1750 / 1944 | 1640 / 1814 | 92.81 |
| 🇺🇸 Los Angeles | DMIT T1 | 818 / 1008 | 493 / 787 | 164.27 |
| 🇺🇸 Los Angeles | GreenCloud | 835 / 1097 | 671 / 803 | 172.60 |
| 🇺🇸 Los Angeles | AlphaVPS | 810 / 999 | 868 / 1049 | 174.48 |
| 🇺🇸 Los Angeles | BAGE LA1 | 806 / 997 | 710 / 1046 | 174.29 |
| 🇺🇸 Los Angeles | BAGE LA2 | 792 / 995 | 672 / 805 | 170.00 |
| 🇺🇸 Los Angeles | RFCHOST T1 | 632 / 814 | 624 / 823 | 172.65 |
| 🇺🇸 San Jose | HostDZire | 851 / 1087 | 810 / 1032 | 174.91 |
| 🇺🇸 San Jose | GreenCloud | 800 / 1088 | 668 / 800 | 173.53 |
| 🇺🇸 San Jose | Nube | 361 / 460 | 589 / 717 | 184.51 |
| 🇺🇸 Seattle | Misaka | 1023 / 1273 | 1022 / 1202 | 149.18 |
| 🇺🇸 Seattle | AlphaVPS | 1015 / 1193 | 980 / 1202 | 149.60 |
| 🇺🇸 Ogden | GreenCloud | 821 / 1008 | 784 / 996 | 179.20 |
| 🇺🇸 Salt Lake City | BAGE | 790 / 1006 | 695 / 974 | 183.74 |
| 🇺🇸 Dallas | AlphaVPS | 404 / 630 | 489 / 749 | 203.18 |
| 🇺🇸 Jacksonville | GreenCloud | 635 / 819 | 604 / 807 | 219.50 |
| 🇺🇸 Chicago | GreenCloud DC1 | 614 / 905 | 713 / 899 | 201.74 |
| 🇺🇸 Chicago | GreenCloud DC2 | 700 / 907 | 705 / 884 | 201.98 |
| 🇺🇸 Ashburn | GreenCloud | 525 / 727 | 684 / 870 | 204.47 |
| 🇺🇸 New York | AlphaVPS | 523 / 812 | 605 / 785 | 227.34 |
| 🇷🇺 Moscow | Misaka | 722 / 998 | 702 / 911 | 184.33 |
| 🇩🇪 Frankfurt | BAGE | 977 / 1206 | 450 / 617 | 149.10 |
| 🇩🇪 Frankfurt | GreenCloud | 977 / 1202 | 316 / 485 | 148.87 |
| 🇩🇪 Frankfurt | Nube | 438 / 769 | 499 / 697 | 193.74 |
| 🇩🇪 Frankfurt | BERO | 815 / 1001 | 676 / 1008 | 158.26 |
| 🇩🇪 Nuremberg | Netcup | 529 / 728 | 366 / 708 | 245.78 |
| 🇩🇪 Nuremberg | AlphaVPS | 724 / 1031 | 746 / 899 | 156.51 |
| 🇳🇱 Amsterdam | GreenCloud | 657 / 904 | 975 / 1165 | 158.31 |
| 🇨🇭 Zurich | Aluy | 1009 / 1289 | 937 / 1117 | 146.52 |
| 🇧🇬 Sofia | AlphaVPS | 239 / 335 | 755 / 930 | 182.95 |
| 🇬🇧 London | AlphaVPS | 947 / 1179 | 755 / 893 | 153.60 |
| 🇬🇧 Coventry | BAGE | 376 / 581 | 450 / 660 | 160.95 |
| 🇬🇧 Coventry | BAGE | 977 / 1198 | 748 / 884 | 155.53 |
🇸🇬 DC2
| Region | Node | Average/max upload (Mbps) | Average/max download (Mbps) | Latency (ms) |
|---|---|---|---|---|
| 🇸🇬 Singapore | RFCHOST T1 | 7885 / 7995 | 5884 / 6134 | 1.19 |
| 🇸🇬 Singapore | BAGE SG | 3781 / 3817 | 4115 / 4949 | 0.86 |
| 🇸🇬 Singapore | BAGE SGS | 3901 / 4100 | 4566 / 5092 | 1.57 |
| 🇸🇬 Singapore | Nube | 5609 / 5805 | 11 / 23 | 1.60 |
| 🇸🇬 Singapore | IDC.Care | 903 / 996 | 901 / 1055 | 1.62 |
| 🇭🇰 Hong Kong | DMIT T1 | 2674 / 2870 | 3249 / 3491 | 30.54 |
| 🇭🇰 Hong Kong | RFCHOST JINX | 812 / 999 | 700 / 992 | 179.40 |
| 🇭🇰 Hong Kong | RFCHOST CTC | 1788 / 2279 | 3072 / 3720 | 39.39 |
| 🇭🇰 Hong Kong | Nube | 1883 / 2186 | 5186 / 5545 | 33.52 |
| 🇭🇰 Hong Kong | GreenCloud DC1 | 5634 / 5847 | 5248 / 5642 | 30.47 |
| 🇭🇰 Hong Kong | GreenCloud DC2 | 4910 / 5305 | 3994 / 4766 | 38.86 |
| 🇭🇰 Hong Kong | HostHatch | 5482 / 5738 | 4253 / 4418 | 30.31 |
| 🇭🇰 Hong Kong | YxVM VOL | 1363 / 1648 | 4015 / 4409 | 39.63 |
| 🇭🇰 Hong Kong | FxTransit | 1821 / 2305 | 4137 / 4421 | 39.79 |
| 🇭🇰 Hong Kong | Hytron | 1656 / 2195 | 4240 / 4534 | 33.44 |
| 🇭🇰 Hong Kong | Halo STD | 1777 / 2012 | 3394 / 3743 | 33.87 |
| 🇭🇰 Hong Kong | Halo Lite | 2122 / 2409 | 3403 / 3731 | 38.10 |
| 🇭🇰 Hong Kong | BAGE HKS | 978 / 1036 | 995 / 1088 | 39.71 |
| 🇭🇰 Hong Kong | BAGE HKL | 980 / 1034 | 1005 / 1098 | 33.55 |
| 🇭🇰 Hong Kong | ClawCloud | 998 / 1096 | 1031 / 1053 | 38.54 |
| 🇭🇰 Hong Kong | Servps | 932 / 1001 | 1073 / 1423 | 34.84 |
| 🇭🇰 Hong Kong | Boil HKT | 890 / 985 | 906 / 965 | 37.55 |
| 🇭🇰 Hong Kong | Kaze HKBN | 1100 / 1216 | 2149 / 2253 | 32.03 |
| 🇹🇼 Taipei | Misaka | 1566 / 1902 | 1847 / 2139 | 70.26 |
| 🇹🇼 Taichung | HiNet | 695 / 818 | 743 / 778 | 58.35 |
| 🇯🇵 Tokyo | DMIT T1 | 1908 / 2056 | 2422 / 2617 | 66.77 |
| 🇯🇵 Tokyo | BAGE JP | 2330 / 2640 | 2151 / 2322 | 67.45 |
| 🇯🇵 Tokyo | BAGE JPS | 2418 / 2643 | 638 / 829 | 214.67 |
| 🇯🇵 Tokyo | Nube | 1548 / 2270 | 1754 / 1892 | 72.82 |
| 🇯🇵 Tokyo | GreenCloud SB | 1604 / 1859 | 2096 / 2300 | 76.27 |
| 🇯🇵 Tokyo | GreenCloud IIJ | 34 / 272 | 1448 / 1756 | 78.71 |
| 🇯🇵 Tokyo | RFCHOST T1 | 628 / 813 | 499 / 632 | 219.50 |
| 🇯🇵 Tokyo | YxVM VOL | 2195 / 2447 | 630 / 811 | 70.51 |
| 🇯🇵 Tokyo | FxTransit | 118 / 189 | 611 / 811 | 71.59 |
| 🇰🇷 Seoul | AWS Lightsail | 1645 / 2079 | 1785 / 1919 | 85.32 |
| 🇻🇳 Hanoi | GreenCloud | 3127 / 3363 | 2029 / 2166 | 61.72 |
| 🇻🇳 Ho Chi Minh City | GreenCloud | 4132 / 4803 | 3641 / 3909 | 34.59 |
| 🇮🇳 Mumbai | HostDZire | 440 / 485 | 388 / 480 | 55.13 |
| 🇦🇺 Sydney | GreenCloud | 1317 / 1553 | 1560 / 1720 | 92.60 |
| 🇺🇸 Los Angeles | DMIT T1 | 629 / 816 | 734 / 950 | 166.62 |
| 🇺🇸 Los Angeles | GreenCloud | 856 / 1092 | 680 / 812 | 168.61 |
| 🇺🇸 Los Angeles | AlphaVPS | 882 / 1098 | 874 / 1054 | 168.70 |
| 🇺🇸 Los Angeles | BAGE LA1 | 840 / 1082 | 730 / 1031 | 166.30 |
| 🇺🇸 Los Angeles | BAGE LA2 | 787 / 1066 | 677 / 813 | 167.30 |
| 🇺🇸 Los Angeles | RFCHOST T1 | 632 / 820 | 664 / 799 | 175.99 |
| 🇺🇸 San Jose | HostDZire | 883 / 1097 | 840 / 1026 | 170.39 |
| 🇺🇸 San Jose | GreenCloud | 823 / 999 | 618 / 765 | 181.23 |
| 🇺🇸 San Jose | Nube | 425 / 656 | 797 / 972 | 189.97 |
| 🇺🇸 Seattle | Misaka | 925 / 1179 | 899 / 1078 | 161.65 |
| 🇺🇸 Seattle | AlphaVPS | 830 / 1001 | 837 / 1022 | 177.34 |
| 🇺🇸 Ogden | GreenCloud | 576 / 812 | 570 / 753 | 238.21 |
| 🇺🇸 Salt Lake City | BAGE | 576 / 811 | 582 / 770 | 238.60 |
| 🇺🇸 Dallas | AlphaVPS | 641 / 813 | 515 / 655 | 201.74 |
| 🇺🇸 Jacksonville | GreenCloud | 581 / 812 | 563 / 744 | 230.04 |
| 🇺🇸 Chicago | GreenCloud DC1 | 489 / 720 | 434 / 567 | 241.84 |
| 🇺🇸 Chicago | GreenCloud DC2 | 633 / 813 | 613 / 797 | 208.42 |
| 🇺🇸 Ashburn | GreenCloud | 504 / 816 | 620 / 806 | 222.62 |
| 🇺🇸 New York | AlphaVPS | 602 / 814 | 558 / 737 | 230.46 |
| 🇷🇺 Moscow | Misaka | 403 / 523 | 725 / 900 | 175.59 |
| 🇩🇪 Frankfurt | BAGE | 924 / 1096 | 722 / 1099 | 161.84 |
| 🇩🇪 Frankfurt | GreenCloud | 586 / 815 | 400 / 566 | 242.55 |
| 🇩🇪 Frankfurt | Nube | 462 / 907 | 623 / 940 | 196.98 |
| 🇩🇪 Frankfurt | BERO | 960 / 1186 | 743 / 888 | 155.99 |
| 🇩🇪 Nuremberg | Netcup | 442 / 630 | 506 / 721 | 248.84 |
| 🇩🇪 Nuremberg | AlphaVPS | 814 / 1154 | 938 / 1136 | 156.77 |
| 🇳🇱 Amsterdam | GreenCloud | 706 / 907 | 660 / 805 | 167.59 |
| 🇨🇭 Zurich | Aluy | 940 / 1181 | 723 / 857 | 160.34 |
| 🇧🇬 Sofia | AlphaVPS | 369 / 537 | 614 / 755 | 182.12 |
| 🇬🇧 London | AlphaVPS | 780 / 1086 | 895 / 1084 | 166.87 |
| 🇬🇧 Coventry | BAGE | 351 / 545 | 541 / 678 | 192.01 |
| 🇬🇧 Coventry | BAGE | 696 / 999 | 848 / 1031 | 171.29 |
Hong Kong-Singapore routes
DMIT, YxVM, and Halo (Akari) all use GSL both ways with DC1. RFC outbound uses the CTC internal network and returns via RETN.
| 🇭🇰 DMIT <-> 🇸🇬 DC1 | 🇭🇰 RFCHOST JINX <-> 🇸🇬 DC1 | 🇭🇰 YxVM <-> 🇸🇬 DC1 | 🇭🇰 Halo STD <-> 🇸🇬 DC1 |
|---|---|---|---|
| Outbound | Outbound | Outbound | Outbound |
| Return | Return | Return | Return |
DMIT and DC2 use GSL both ways and have the lowest latency. JINX to DC2 detours and has very high latency. YxVM and DC2 use Leaseweb backbone through HKIX. Halo (Akari) outbound uses COGENT->TATA and returns through the Leaseweb backbone.
| 🇭🇰 DMIT <-> 🇸🇬 DC2 | 🇭🇰 RFCHOST JINX <-> 🇸🇬 DC2 | 🇭🇰 YxVM <-> 🇸🇬 DC2 | 🇭🇰 Halo STD <-> 🇸🇬 DC2 |
|---|---|---|---|
| Outbound | Outbound | Outbound | Outbound |
| Return | Return | Return | Return |
7. Summary
Considering Singapore landing nodes are mostly pulled from Hong Kong, DC1 has a much clearer advantage and lower latency. If the Hong Kong side is DMIT/GSL, both are acceptable.
Neither is suitable for direct connection, especially DC2. But DC1 Mobile direct connection is not great either.
Unlocking is similar for both and depends on the IP range; currently it is quite good. GreenCloud does not guarantee unlocking and does not lock geolocation, so it may get worse once more people use it.
8. 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
9. Acknowledgements
Thanks to “Impart Speed Test“(@Kyoma) for providing public-interest speed-test service.