0. Preface
Claw’s international route has always had Hong Kong nodes; the ones previously removed were Hong Kong nodes on the China-optimized route. Let’s test how this one performs 
1. Basic Information
| Plan | 🇭🇰 ClawCloud Cloud VDS 4GB |
|---|---|
| Location | China, Hong Kong |
| CPU | 1 core / 2 threads x vCPU Intel Xeon Platinum |
| RAM | 4GB |
| Disk | 80GB SSD |
| Transfer | 1000GB(out) |
| Speed | 1Gbps |
| Network | 1 x IPv4 + 1 x /128 IPv6 |
| Price | Monthly USD$4.20(original price USD$12.00) |
Includes five manual backups and 30 days of daily automatic backups
Note: Hong Kong nodes on the international route may still have drop/throttling issues. SeePangge’s YouTube video
2. YABS
Judging from the single-core/multi-core scores, hyper-threading is probably being used
curl -sL https://yabs.sh | bash -s -- -i -9 -6
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2025-04-20 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Sun Sep 28 03:50:37 PM CST 2025
Basic System Information:
---------------------------------
Uptime : 0 days, 0 hours, 50 minutes
Processor : Intel(R) Xeon(R) Platinum
CPU cores : 2 @ 2499.998 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 3.7 GiB
Swap : 0.0 KiB
Disk : 78.2 GiB
Distro : Debian GNU/Linux 13 (trixie)
Kernel : 6.12.43+deb13-cloud-amd64
VM Type : KVM
IPv4/IPv6 : ✔ Online / ✔ Online
IPv6 Network Information:
---------------------------------
ISP : Alibaba (US) Technology Co., Ltd.
ASN : AS45102 Alibaba (US) Technology Co., Ltd.
Host : Alibaba.com LLC
Location : Hong Kong, Kowloon ()
Country : Hong Kong
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 20.68 MB/s (5.1k) | 66.01 MB/s (1.0k)
Write | 20.69 MB/s (5.1k) | 66.40 MB/s (1.0k)
Total | 41.37 MB/s (10.3k) | 132.42 MB/s (2.0k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 63.44 MB/s (123) | 62.72 MB/s (61)
Write | 66.84 MB/s (130) | 67.32 MB/s (65)
Total | 130.28 MB/s (253) | 130.04 MB/s (126)
Geekbench 4 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 5188
Multi Core | 5605
Full Test | https://browser.geekbench.com/v4/cpu/19048117
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1198
Multi Core | 1362
Full Test | https://browser.geekbench.com/v5/cpu/23814364
Geekbench 6 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 1500
Multi Core | 1740
Full Test | https://browser.geekbench.com/v6/cpu/14126937
3. NodeQuality
🎬 IP Quality
A typical Alibaba IP: streaming does not unlock, and websites trigger verification
| IPv4 | IPv6 |
|---|---|
🌐 Network Quality
All three Chinese networks route indirectly with no direct path at all; it does not even connect to Lumen
| IPv4 | IPv6 |
|---|---|
📍 Return Routes
| IPv4 | IPv6 |
|---|---|
4. nws.sh international speed test
curl -sL nws.sh | bash
---------------------------------- nws.sh ---------------------------------
A simple script to bench network performance using speedtest-cli
---------------------------------------------------------------------------
Version : v2025.08.30
Global Speedtest : wget -qO- nws.sh | bash
Region Speedtest : wget -qO- nws.sh | bash -s -- -r <region>
Ping & Routing : wget -qO- nws.sh | bash -s -- -rt <region>
-----------------------------------------------------------------------------------------------------------------------
Speedtest.net (Region: GLOBAL)
---------------------------------------------------------------------------
Location Latency Loss DL Speed UP Speed Server
ISP: Alibaba
Nearest 2.51 ms N/A 998.47 Mbps 955.60 Mbps Misaka Network, Inc. - Hong Kong
Bangalore, IN 92.15 ms 0.0% 1005.70 Mbps 983.94 Mbps Bharti Airtel Ltd - Bangalore
Chennai, IN FAILED
Mumbai, IN 86.15 ms 0.0% 1027.94 Mbps 1007.19 Mbps Melbicom - Mumbai
Seattle, US 181.44 ms N/A 1009.41 Mbps 477.15 Mbps Comcast - Seattle, WA
Los Angeles, US 149.63 ms 0.0% 995.52 Mbps 540.20 Mbps ReliableSite Hosting - Los Angeles, CA
Dallas, US 183.42 ms 0.0% 479.50 Mbps 458.77 Mbps Hivelocity - Dallas, TX
Miami, US 209.17 ms 0.0% 1000.25 Mbps 387.14 Mbps Telxius - Miami, FL
New York, US 188.26 ms 0.0% 1034.46 Mbps 438.25 Mbps GSL Networks - New York, NY
Toronto, CA 222.42 ms 0.0% 1024.50 Mbps 363.97 Mbps Rogers - Toronto, ON
Mexico City, MX 184.82 ms N/A 1017.33 Mbps 437.87 Mbps INFINITUM - Ciudad de México
London, UK 188.39 ms 0.0% 1039.30 Mbps 421.57 Mbps VeloxServ Communications - London
Amsterdam, NL 243.90 ms 0.0% 1027.21 Mbps 339.92 Mbps 31173 Services AB - Amsterdam
Paris, FR FAILED
Frankfurt, DE 276.61 ms 0.0% 1021.09 Mbps 275.95 Mbps Clouvider Ltd - Frankfurt am Main
Warsaw, PL 256.84 ms 0.0% 1000.98 Mbps 284.10 Mbps Play - Warszawa
Bucharest, RO 205.10 ms 0.0% 1022.95 Mbps 409.35 Mbps Vodafone Romania Mobile - Bucharest - Bucharest
Moscow, RU FAILED
Jeddah, SA 224.42 ms 0.0% 1039.44 Mbps 359.19 Mbps Saudi Telecom Company
Dubai, AE 106.04 ms N/A 1056.62 Mbps 777.88 Mbps e& UAE - Dubai
Istanbul, TR 335.18 ms 0.0% 999.96 Mbps 228.50 Mbps Turkcell - Istanbul
Tehran, IR 252.27 ms 0.0% 927.57 Mbps 326.12 Mbps Irancell - Tehran
Cairo, EG 272.13 ms 0.3% 1031.24 Mbps 139.87 Mbps Telecom Egypt - Cairo
Tokyo, JP 51.82 ms 3.0% 1000.46 Mbps 1051.47 Mbps fdcservers.net - Tokyo
Shanghai, CU-CN 137.78 ms N/A 921.19 Mbps 0.23 Mbps China Unicom 5G - Shanghai
Hong Kong, CN 1.87 ms N/A 998.07 Mbps 454.87 Mbps Misaka Network, Inc. - Hong Kong
Singapore, SG 37.55 ms 0.0% 1013.84 Mbps 1051.42 Mbps ViewQwest - Singapore
Jakarta, ID 117.04 ms 0.0% 1.73 Mbps 890.21 Mbps PT Solnet Indonesia - Jakarta
Sydney, AU 137.39 ms 0.0% 1021.25 Mbps 632.13 Mbps Aussie Broadband - Sydney
---------------------------------------------------------------------------
Avg DL Speed : 950.62 Mbps
Avg UL Speed : 526.65 Mbps
Total DL Data : 35.60 GB
Total UL Data : 18.58 GB
Total Data : 54.18 GB
---------------------------------------------------------------------------
Duration : 13 min 22 sec
System Time : 28/09/2025 - 15:20:37 CST
Total Script Runs : 409263
---------------------------------------------------------------------------
Result : https://result.nws.sh/r/1759043283_RE34FK_GLOBAL.txt
---------------------------------------------------------------------------
5. iPerf3 TCP single-thread speed test (single 30-second test, 2025-09-28 18:00)
Full speed within Hong Kong; Japan and Singapore are still decent. DMIT JP probably has a firewall enabled, limiting uploads
| Peer | Download from peer(Mbps) | Retransmits | Upload to peer(Mbps) | Retransmits | Average latency(ms) |
|---|---|---|---|---|---|
| 🇭🇰 Hong Kong DMIT T1 @4Gbps | 1000 | 240663 | 1040 | 30810 | 3.380 |
| 🇭🇰 Hong Kong RFC T1 JINX | 1000 | 252245 | 1030 | 56078 | 2.154 |
| 🇭🇰 Hong Kong RFC T1 CTC | 1010 | 9245 | 1040 | 31564 | 2.646 |
| 🇭🇰 Hong Kong BAGE HKS @1Gbps | 951 | 6149 | 989 | 0 | 2.359 |
| 🇭🇰 Hong Kong BAGE HKL @1Gbps | 1010 | 240664 | 981 | 0 | 2.263 |
| 🇭🇰 Hong Kong SerVPS SP @1Gbps | 998 | 253600 | 934 | 0 | 3.003 |
| 🇭🇰 Hong Kong VPS.Town B1 @1Gbps | 908 | 0 | 933 | 0 | 3.003 |
| 🇭🇰 Hong Kong Boil HKBN @2.5Gbps | 1000 | 240164 | 1030 | 46861 | 2.153 |
| 🇹🇼 Taichung Boil HiNet @777Mbps | 686 | 854 | 700 | 1334 | 34.804 |
| 🇯🇵 Tokyo DMIT T1 @4Gbps | 327 | 17 | 9.82 | 5695 | 49.089 |
| 🇯🇵 Tokyo RFC JP1T1 | 436 | 0 | 431 | 0 | 51.039 |
| 🇯🇵 Tokyo BAGE JPS @5Gbps | 674 | 0 | 425 | 0 | 50.367 |
| 🇯🇵 Tokyo DDPS @1Gbps | 433 | 17 | 432 | 5 | 54.907 |
| 🇯🇵 Tokyo GGV GSL @1Gbps | 616 | 0 | 412 | 289 | 52.687 |
| 🇯🇵 Tokyo AWS Lightsail @1Gbps | 306 | 179 | 344 | 0 | 52.476 |
| 🇸🇬 Singapore RFC T1 | 595 | 0 | 515 | 0 | 42.614 |
| 🇸🇬 Singapore BAGE GSL @5Gbps | 546 | 45 | 556 | 29 | 40.612 |
| 🇸🇬 Singapore BAGE SGS @5Gbps | 603 | 2 | 607 | 0 | 32.118 |
| 🇸🇬 Singapore IDC.Care @1Gbps | 496 | 179 | 585 | 0 | 42.454 |
| 🇸🇬 Singapore AWS Lightsail @1Gbps | 482 | 0 | 479 | 0 | 34.049 |
| 🇺🇸 Los Angeles DMIT T1 @4Gbps | 130 | 0 | 136 | 0 | 153.291 |
| 🇺🇸 Los Angeles RFC T1 | 114 | 0 | 111 | 0 | 168.137 |
| 🇺🇸 Los Angeles BAGE LA1 @1Gbps | 127 | 127 | 121 | 855 | 156.404 |
| 🇺🇸 Los Angeles BAGE LA2 @1Gbps | 124 | 0 | 137 | 255 | 161.549 |
| 🇺🇸 Los Angeles BAGE LA3 @1Gbps | 144 | 0 | 136 | 0 | 146.763 |
| 🇺🇸 San Francisco HostDZire @10Gbps | 129 | 0 | 131 | 0 | 157.088 |
| 🇺🇸 Salt Lake City BAGE @1Gbps | 130 | 0 | 130 | 0 | 157.449 |
| 🇩🇪 Frankfurt BAGE @5Gbps | 119 | 0 | 119 | 0 | 174.065 |
| 🇩🇪 Nuremberg Netcup @1Gbps | 102 | 0 | 78.1 | 0 | 251.167 |
| 🇬🇧 Coventry BAGE @1Gbps | 105 | 0 | 92.1 | 275 | 190.433 |
| 🇮🇳 Mumbai HostDZire @10Gbps | 226 | 0 | 180 | 1404 | 94.428 |
6. HTTP website speed test (itdog.cn)
| No CDN | Cloudflare CDN | China-optimized CDN |
|---|---|---|
7. Summary
| Pros | Cons |
|---|---|
| Daily automatic backups + five manual backups | Alibaba IPs are very dirty; unsuitable for exit-node unlocking |
| Less likely to be hit by IP-range scans | If actually attacked, nullrouting lasts 24+ hours |
| International connectivity is acceptable | Return to China routes indirectly on all three networks |
| 1Gbps large bandwidth | Has a history of throttling |
It is more suitable for hosting a small personal site or probe panel behind a CDN; the automatic backups are nice. Since it is backed by Alibaba, it should be less affected by DDoS range scanning. But given its past throttling history, be cautious about production use. It is not suitable as an exit node either, because Alibaba IPs are very dirty. Do not consider it for return-to-China use; all three networks route indirectly, worse than direct West Coast US routes
8. About The Tests
All machines used in this article were purchased at my own expense; TCP tuning was limited to enabling BBR+FQ, with all other parameters left at defaults; results may be time-sensitive