Geelinx, an IDC provider founded by students studying in the United Kingdom and Japan, currently sells low-cost VPS in Japan, the United Kingdom, and the Netherlands. Its Japan region uses Netlab. Although that upstream has been widely criticized, Geelinx's friendly pricing has still made it quite popular.
I previously wrote a launch review of Geelinx. Recently I found that, whether in the Geelinx customer group or in my own tests, the JP product's single-thread return-to-China speed often cannot reach the advertised bandwidth, being limited to around 50Mbps (though there are also reports in the Geelinx customer group of no speed limit), while multi-thread tests do not show this phenomenon.I do not know whether this is due to upstream Netlab policy or Geelinx's own limit (Update: according to the provider, they were deceived, and sold 50M machines ). I also do not know whether the limit is by region or by IP/host. Expectations for so-called "route machines" should be lowered further. After all, the provider itself said this is not a route machine. For now, it is still okay for unlocking Netflix and Disney+ and serving as a backup for Unicom and Telecom users. 
(1) Network quality
Speed: across multiple regional carriers (five locations in the Yangtze River Delta and Pearl River Delta), single-thread return-to-China speeds are limited below 50Mbps, except Shanghai Telecom in the test, which can reach 200Mbps+ both day and night. Telecom basically runs right against the 50M single-thread limit. Unicom and Mobile are similar but slightly slower than Telecom; Guangzhou Unicom and Foshan Mobile in the test are QoS-limited to Kbps levels during the evening peak. Multi-thread can break through this limit, but still cannot reach the advertised bandwidth. The improvement over before is that the previous cutout issue did not appear, and all tests completed smoothly.
Latency: Telecom and Unicom latency increases slightly during the evening peak. Overall packet loss and jitter are acceptable, though a few regions have packet loss and jitter both day and night. Mobile latency is higher than Telecom/Unicom, and it jitters severely during the evening peak; NetQuality also measured fairly serious TCP ping packet loss in most provinces and cities.
Routes: return routes are unchanged: Telecom 163, Unicom 10099, Mobile CMI. Outbound uses each network's own direct backbone route, all entering NTT; compared with launch, this is a downgrade. Previously I measured Telecom CN2->NTT, Unicom 4837 direct, and Mobile through Lumen.
International: local landing nodes can almost saturate 300Mbps, and US West can also nearly saturate. Seattle latency is 88 ms+, Los Angeles is 99 ms+. IPv6 quality is still poor; even pinging local v6 gives 60 ms+ latency. Cloudflare speed-test nodes may even be assigned to Taiwan, doubling latency, and speed is worse than v4. I recommend disabling v6 or changing outbound priority to v4.
(2) IP quality
Anycast/broadcast IP. IPQuality measured full IPv4 international streaming unlock and correct location, but JP local streaming is basically not unlocked. IPv6 is from Geelinx's own AS, with poor unlock and confusing geolocation: Google places it in the UK, Bing and Apple in the US, so latency is high.
(3) Machine performance
The CPU is a domestic Zhaoxin, roughly similar to E5. Single-core is mediocre, while the doubled launch multi-core is still decent. Geekbench 5 scores are 640 single-core and 2408 multi-core, down 12.3% and 14.8% from the launch test respectively. I/O has not changed much and remains fairly poor.
All tested products were purchased at my own expense; this article has no affiliate links.
Product Configuration
| Plan | 🇯🇵 GeeLinx JP.TYO.BGP-A-P |
|---|---|
| Location | Japan, Tokyo |
| CPU | 4 x Zhaoxin KH-4000 vCores |
| RAM | 4GB |
| Disk | 20GB |
| Traffic | 800GB(outbound billing) |
| Peak Bandwidth | 300Mbps |
| Return Route | Telecom 163/Unicom10099/Mobile CMI |
| Outbound Route | Telecom 163/Unicom 4837/Mobile CMI |
| Network | 1 x IPv4, 1 x /64 IPv6 |
| Price | Annual EUR€16.88 |
Provider website: https://www.geelinx.com
Test Overview
| Live Probe | YABS | Node Quality | Mixed Benchmark | nws.sh | Streaming Unlock |
|---|---|---|---|---|---|
| [Link] | [Link] | [Link] | [Link] | [Link] | [Link] |
1. Network Quality
Single-Thread Return-to-China Speed Tests
| Evening Peak | Off-Peak |
|---|---|
Single-thread & Multi-thread Speed Test Comparison
| Single-thread | Multi-thread |
|---|---|
NetQuality
bash ‹(curl -sL https://Check. Place) -N
| Evening Peak | Off-Peak |
|---|---|
Evening Peak Outbound TCP Ping
Test Site: itdog.cn
Test Time: 2026-03-10 22:00
| China Telecom | China Unicom | China Mobile |
|---|---|---|
Off-Peak Outbound TCP Ping
Test Site: itdog.cn
Test Time: 2026-03-10 10:00
| China Telecom | China Unicom | China Mobile |
|---|---|---|
Return Routes
bash ‹(curl -sL https://Check. Place) -N
Outbound Routes
Test Site: itdog.cn
Test Time: 2026-03-12 17:00
| ShanghaiTelecom | ShanghaiUnicom | Shanghai Mobile |
|---|---|---|
bgp.tools
Test Site: bgp.tools
Test Time: 2025-03-12 17:00
| IPv4 | IPv6 |
|---|---|
Web Latency
bash <(wget -qO- https://raw.githubusercontent.com/danger-dream/network-latency-tester/main/latency.sh)
| IPv4 Preferred | IPv6 Preferred |
|---|---|
International Latency(iplark)
Test Site: iplark.com
Test Time: 2025-03-12 17:00
| IPv4 | IPv6 |
|---|---|
Own-node Tests
Test Method: iPerf3 TCP single-thread & ICMP Ping, one 10-second run
Test Time: 2025-03-12 17:00
IPv4
| iPerf3 | ICMP Ping |
|---|---|
IPv6
| iPerf3 | ICMP Ping |
|---|---|
Cloudflare Local Speed Test
wget -qO- https://github.com/code-inflation/cfspeedtest/releases/latest/download/cfspeedtest-x86_64-unknown-linux-gnu.tar.gz | tar xz && ./cfspeedtest -n 25; rm ./cfspeedtest
v6 was assigned to a Taiwan speed-test node.
IPv4
Avg GET request latency 44.16 ms (RTT excluding server processing time)
Type Payload | min/max/avg in mbit/s
Download 100KB | min 6.37 max 25.74 avg 17.43
Download 1MB | min 48.00 max 146.34 avg 114.48
Download 10MB | min 200.06 max 260.02 avg 249.53
Upload 100KB | min 20.95 max 57.47 avg 46.45
Upload 1MB | min 84.68 max 249.93 avg 189.78
Upload 10MB | min 164.03 max 279.26 avg 253.91
IPv6
Avg GET request latency 92.75 ms (RTT excluding server processing time)
Type Payload | min/max/avg in mbit/s
Download 100KB | min 3.16 max 9.01 avg 6.78
Download 1MB | min 36.33 max 81.16 avg 67.89
Download 10MB | min 17.91 max 226.20 avg 126.21
Upload 100KB | min 4.77 max 15.19 avg 11.04
Upload 1MB | min 21.59 max 122.64 avg 81.08
Upload 10MB | min 81.24 max 185.07 avg 153.35
Apple Local Speed Test
curl -sL nxtrace.org/speedtest | bash
▸ Idle Latency
[+] Endpoint: https://mensura.cdn-apple.com/api/v1/gm/small
[+] Samples: 20
➜ 103.95 ms median (min 99.62 / avg 105.08 / max 117.73) jitter 0.95 ms
▸ Download (single thread)
[+] Threads: 1
[+] Limit: 2G / 10s per thread
➜ 273 Mbps (329.0 MiB in 10.1s)
[+] Loaded latency: 104.28 ms (jitter 0.38 ms)
▸ Download (multi-thread)
[+] Threads: 4
[+] Limit: 2G / 10s per thread
➜ 276 Mbps (331.8 MiB in 10.1s, 4 threads)
[+] Loaded latency: 103.61 ms (jitter 0.59 ms)
▸ Upload (single thread)
[+] Threads: 1
[+] Limit: 2G / 10s per thread
➜ 131 Mbps (156.1 MiB in 10.0s)
[+] Loaded latency: 103.89 ms (jitter 0.43 ms)
▸ Upload (multi-thread)
[+] Threads: 4
[+] Limit: 2G / 10s per thread
➜ 254 Mbps (306.2 MiB in 10.1s, 4 threads)
[+] Loaded latency: 113.15 ms (jitter 1.09 ms)
Speedtest International Test
curl -sL nws.sh | bash
---------------------------------- nws.sh ---------------------------------
A simple script to bench network performance using speedtest-cli
---------------------------------------------------------------------------
Version : v2025.11.07
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: NetLab Global
Nearest 1.14 ms 1.4% 286.56 Mbps 279.01 Mbps サーバーならDDPS.jp - Tokyo
Bangalore, IN 187.84 ms 0.0% 150.96 Mbps 484.87 Mbps Bharti Airtel Ltd - Bangalore
Chennai, IN 165.81 ms 0.0% 94.16 Mbps 470.47 Mbps RailTel Corporation of India Ltd - Chennai
Mumbai, IN 193.32 ms 0.0% 83.75 Mbps 585.58 Mbps Melbicom - Mumbai
Seattle, US 167.57 ms N/A 94.01 Mbps 405.35 Mbps Comcast - Seattle, WA
Los Angeles, US 175.29 ms 0.0% 113.81 Mbps 372.07 Mbps ReliableSite Hosting - Los Angeles, CA
Dallas, US 269.36 ms 0.0% 98.37 Mbps 522.71 Mbps Hivelocity - Dallas, TX
Miami, US 243.39 ms 0.0% 101.35 Mbps 447.31 Mbps Frontier - Miami, FL
New York, US 146.90 ms 0.0% 313.95 Mbps 282.78 Mbps GSL Networks - New York, NY
Toronto, CA 276.13 ms 0.0% 100.00 Mbps 275.83 Mbps Rogers - Toronto, ON
Mexico City, MX FAILED
London, UK 275.97 ms 0.0% 98.75 Mbps 500.76 Mbps VeloxServ Communications - London
Amsterdam, NL FAILED
Paris, FR 307.85 ms 0.0% 98.35 Mbps 289.59 Mbps Scaleway - Paris
Frankfurt, DE FAILED
Warsaw, PL 380.44 ms 23.6% 96.69 Mbps 287.97 Mbps Play - Warszawa
Bucharest, RO 330.76 ms 0.0% 95.07 Mbps 365.05 Mbps Digi
Moscow, RU FAILED
Jeddah, SA 334.87 ms 0.0% 102.01 Mbps 210.53 Mbps Saudi Telecom Company
Dubai, AE 141.06 ms N/A 320.89 Mbps 319.63 Mbps e& UAE - Dubai
Istanbul, TR 346.91 ms 0.0% 90.93 Mbps 238.99 Mbps Turkcell - Istanbul
Tehran, IR FAILED
Cairo, EG 251.89 ms 0.0% 280.25 Mbps 197.17 Mbps Telecom Egypt - Cairo
Tokyo, JP FAILED
Shanghai, CU-CN 74.04 ms N/A 90.80 Mbps 354.84 Mbps China Unicom 5G - Shanghai
Hong Kong, CN FAILED
Singapore, SG 79.03 ms 0.0% 288.01 Mbps 281.01 Mbps ViewQwest - Singapore
Jakarta, ID 84.18 ms 0.0% 285.91 Mbps 276.07 Mbps PT Solnet Indonesia - Jakarta
Sydney, AU 256.12 ms 0.0% 93.61 Mbps 18.93 Mbps Aussie Broadband - Sydney
---------------------------------------------------------------------------
Avg DL Speed : 153.55 Mbps
Avg UL Speed : 339.39 Mbps
Total DL Data : 4.52 GB
Total UL Data : 11.03 GB
Total Data : 15.55 GB
---------------------------------------------------------------------------
Duration : 15 min 11 sec
System Time : 10/03/2026 - 19:26:00 CST
Total Script Runs : 520832
---------------------------------------------------------------------------
Result : https://result.nws.sh/r/1773141963_J1P7ZD_GLOBAL.txt
---------------------------------------------------------------------------
2. IP Quality
IPQuality
bash ‹(curl -sL https://Check. Place) -I
Streaming Unlock
bash <(curl -L -s check.unlock.media)
| IPv4 | IPv6 |
|---|---|
Common IP Database Checks
export noninteractive=true && curl -L https://raw.githubusercontent.com/oneclickvirt/ecs/master/goecs.sh -o goecs.sh && chmod +x goecs.sh && ./goecs.sh install && goecs
--------------------------------------IP Quality Check--------------------------------------
The following are database IDs; results include the corresponding source database IDs
ipinfo Database [0] | scamalytics Database [1] | virustotal Database [2] | abuseipdb Database [3] | ip2location Database [4]
ip-api Database [5] | ipwhois Database [6] | ipregistry Database [7] | ipdata Database [8] | db-ip Database [9]
ipapiis Database [A] | ipapicom Database [B] | bigdatacloud Database [C] | dkly Database [D] | ipqualityscore Database [E]
ipintel Database [F] | ipfighter Database [G] | fraudlogix Database [H] | cloudflare Database [I] |
IPv4:
Security Score:
Trust Score(higher is better): 100 [8]
VPN Score(lower is better): 0 [8]
Proxy Score(lower is better): 0 [8]
Community Votes - Harmless: 0 [2]
Community Votes - Malicious: 0 [2]
Threat Score(lower is better): 0 [8]
Fraud Score(lower is better): 0 [E]
Abuse Score(lower is better): 0 [3 H]
ASN Abuse Score(lower is better): 0 (Very Low) [A]
Company Abuse Score(lower is better): 0 (Very Low) [A]
Threat Level: low [9 B] Low [H]
Traffic Share: Human(higher is better)6% [I] Bot(lower is better)93% [I]
Blacklist Record Statistics (number of blacklist sites with records):
Harmless Records: 0 [2] Malicious Records: 0 [2] Suspicious Records: 0 [2] No-record Count: 94 [2]
Security Info:
Usage Type: residential [H] business [0 7 9] Commercial [3] unknown [C]
Company Type: hosting [7] business [0 A]
Browser Type: Mainstream77% Other22% [I]
Device Type: Desktop96% Mobile3% Other0% [I]
OS Type: Mainstream90% Other9% [I]
Is Cloud Provider: Yes [7 D]
Is Data Center: Yes [A] No [0 5 8 C G H]
Is Mobile Device: No [5 A C G] Yes [E]
Is Proxy: Yes [G] No [0 4 5 7 8 9 A B C D E H]
Is VPN: No [0 7 A C D E G H]
Is Tor: No [0 3 7 8 A B C D E H]
Is Tor Exit: No [7 D]
Is Web Crawler: No [9 A B E]
Is Anonymous: No [7 8 D]
Is Attacker: No [7 8 D]
Is Abuser: No [7 8 A C D E]
Is Threat: No [7 8 C D]
Is Relay: No [0 7 8 C D]
Is Bogon: No [7 8 A C D]
Is Bot: No [E H]
DNS Blacklist: 309(Total_Check) 0(Clean) 0(Blacklisted) 0(Other)
IPv6:
Security Score:
Abuse Score(lower is better): 0 [3 H]
ASN Abuse Score(lower is better): 0 (Very Low) [A]
Company Abuse Score(lower is better): 0 (Very Low) [A]
Threat Level: low [B] Low [H]
Traffic Share: Human(higher is better)6% [I] Bot(lower is better)93% [I]
Security Info:
Usage Type: Commercial [3] residential [H]
Company Type: business [A]
Browser Type: Mainstream77% Other22% [I]
Device Type: Desktop96% Mobile3% Other0% [I]
OS Type: Mainstream90% Other9% [I]
Is Data Center: Yes [A] No [G H]
Is Mobile Device: No [A G]
Is Proxy: No [A B H] Yes [G]
Is VPN: No [A G H]
Is Tor: No [3 A B H]
Is Web Crawler: No [A B]
Is Abuser: No [A]
Is Bogon: No [A]
Is Bot: No [H]
DNS Blacklist: 309(Total_Check) 0(Clean) 0(Blacklisted) 309(Other)
iplark Check Score
Test Site: iplark.com
Test Time: 2025-03-12 17:00
| IPv4 | IPv6 |
|---|---|
3. Machine Performance
HardwareQuality
bash ‹(curl -sL https://Check. Place) -I
YABS
curl -sL https://yabs.sh | bash -s -- -5 -i
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
# Yet-Another-Bench-Script #
# v2025-04-20 #
# https://github.com/masonr/yet-another-bench-script #
# ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## ## #
Thu Mar 12 02:53:26 PM CST 2026
Basic System Information:
---------------------------------
Uptime : 8 days, 23 hours, 50 minutes
Processor : ZHAOXIN KaiSheng KH-40000/32@2GHz
CPU cores : 4 @ 2345.448 MHz
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
RAM : 3.7 GiB
Swap : 0.0 KiB
Disk : 19.5 GiB
Distro : Debian GNU/Linux 13 (trixie)
Kernel : 6.12.63+deb13-cloud-amd64
VM Type : KVM
IPv4/IPv6 : ✔ Online / ✔ Online
IPv6 Network Information:
---------------------------------
ISP : Geelinx LTD
ASN : AS209557 GEELINX LTD
Host : Geelinx LTD
Location : City of London, England (ENG)
Country : United Kingdom
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
---------------------------------
Block Size | 4k (IOPS) | 64k (IOPS)
------ | --- ---- | ---- ----
Read | 7.40 MB/s (1.8k) | 116.48 MB/s (1.8k)
Write | 7.43 MB/s (1.8k) | 117.10 MB/s (1.8k)
Total | 14.84 MB/s (3.7k) | 233.58 MB/s (3.6k)
| |
Block Size | 512k (IOPS) | 1m (IOPS)
------ | --- ---- | ---- ----
Read | 157.26 MB/s (307) | 147.75 MB/s (144)
Write | 165.62 MB/s (323) | 157.59 MB/s (153)
Total | 322.88 MB/s (630) | 305.35 MB/s (297)
Geekbench 5 Benchmark Test:
---------------------------------
Test | Value
|
Single Core | 656
Multi Core | 2515
Full Test | https://browser.geekbench.com/v5/cpu/24173388
YABS completed in 4 min 46 sec
4. Appendix
About the Tests
The machines used in this test were purchased at my own expense; results may be time-sensitive.
Kernel parameters of the test 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
Acknowledgements
Thanks to the administrators and volunteers of Haoyong Aiyong for providing public-interest speed test services; thanks to itdog.cn and iplark.com and the contributors of the test scripts
🔔 I wonder whether everyone's Geelinx JP has encountered speed limiting; feel free to discuss in the comments. Thanks for reading this far. If you want reminders for new NS reviews, you can subscribe to my channel
