HostHatch Seoul, Korea LG Route Black Friday Plan 35 USD/Year | Evening Peak Speed Tests | International Connectivity | Review

I bought a HostHatch Seoul preorder plan on Black Friday, and it was provisioned the day before yesterday, so let's see how it performs.

Contents

1. Configuration
2. Node Quality
3. Mainland Three-Network TCP Ping
4. Evening Peak Direct Single-Thread Speed Tests
5. International Connectivity
6. Summary

1. Configuration

Provider website: https://hosthatch.com

🔗 Quick Links

The upstream connects to MOACK. In the preorder provisioning email, HostHatch said that over the next few days they would add new carriers and optimize network connectivity (Over the next few days, we will be adding new carriers and optimizing connectivity.)

IPv4	IPv6

2. Node Quality

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

💻 Basic Information

Performance is good.

Basic System Information:
---------------------------------
Uptime     : 1 days, 6 hours, 16 minutes
Processor  : AMD EPYC 7763 64-Core Processor
CPU cores  : 2 @ 2449.998 MHz
AES-NI     : ✔ Enabled
VM-x/AMD-V : ✔ Enabled
RAM        : 3.8 GiB
Swap       : 0.0 KiB
Disk       : 46.6 GiB
Distro     : Debian GNU/Linux 12 (bookworm)
Kernel     : 6.12.57+deb13-cloud-amd64
VM Type    : KVM
IPv4/IPv6  : ✔ Online / ✔ Online

IPv6 Network Information:
---------------------------------
ISP        : HostHatch
ASN        : Unknown
Host       : HostHatch, Inc
Location   : Tampa, Florida (FL)
Country    : United States

fio Disk Speed Tests (Mixed R/W 50/50) (Partition -):
---------------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 152.04 MB/s  (38.0k) | 1.72 GB/s    (27.0k)
Write      | 152.44 MB/s  (38.1k) | 1.73 GB/s    (27.1k)
Total      | 304.48 MB/s  (76.1k) | 3.46 GB/s    (54.1k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 4.64 GB/s     (9.0k) | 4.77 GB/s     (4.6k)
Write      | 4.88 GB/s     (9.5k) | 5.09 GB/s     (4.9k)
Total      | 9.53 GB/s    (18.6k) | 9.86 GB/s     (9.6k)

Geekbench 5 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1132                          
Multi Core      | 2216                          
Full Test       | https://browser.geekbench.com/v5/cpu/23976567

 SysBench CPU Test (Fast Mode, 1-Pass @ 5sec)
---------------------------------
 1 Thread Test (Single Core) Score:          3470 Scores
 2 Thread Test (Multi Core) Score:          6931 Scores
 SysBench Memory Test (Fast Mode, 1-Pass @ 5sec)
---------------------------------
 Single-Thread Read Test:          43173.07 MB/s
 Single-Thread Write Test:          25382.72 MB/s

🎬IP Quality

Anycast/broadcast IP; geolocation is not very accurate yet.

IPv4	IPv6

🌐 Network Quality

IPv4	IPv6

📍Return Routes

LG UPLUS route, with direct backbone connectivity for all three mainland networks.

IPv4	IPv6

3. mainland ChinaThree-Network TCP Ping（itdog.cn, 2025-12-16 22:00）

Evening-peak latency is fairly high and jitter is severe, but packet loss is acceptable.

China Telecom	China Unicom	China Mobile

4. Evening Peak Direct Single-Thread Speed Tests(2025-12-16 21:00-23:00)

During the evening peak, China Telecom and China Mobile have almost no speed; China Unicom can still run a bit, though with higher latency.

Telecom

[Telecom Speed Test Record]

Unicom

[Unicom Speed Test Record]

Mobile

[Mobile Speed Test Record]

Premium Network

Shanghai Telecom CN2 is quite good.

[Premium NetworkSpeed Test Record]

5. International Connectivity

Looking at web latency, many sites route through Japan, probably because many websites have no CDN nodes in Korea but do have them in Japan.

HostHatch has plenty of local bandwidth; it saturated the Seoul Lightnode port, and performance to Hong Kong and Japan is also good, though latency still needs optimization.

Web Latency

bash <(wget -qO- https://raw.githubusercontent.com/danger-dream/network-latency-tester/main/latency.sh)

Latency Test（ping.pe)

IPv4	IPv6

Speed Test - iPerf3 TCP IPv4 Single-Thread (single 10-second run, 2025-12-15)

Test Tool: iPerf3

[Test Record]

6. Summary

Performance is fairly good and it suits websites aimed at Japan and Korea, but it is not very suitable for proxy use. Although the route back to China is LG UPLUS, evening-peak direct speed still cannot climb. At the same time, accessing many overseas sites from Korea, especially global services, often requires detouring through Japan, adding more than 30 ms of latency. For pure proxy use, if you do not strictly need a Korea landing node, a Japan landing node is better; Japan has more mature direct optimized routes and dedicated-line resources, with more choices.

7. About the Tests

The machines used in this test, including all machines used for iPerf3 speed tests, 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

8. Acknowledgements

Thanks to "Impart Speed Test“(@Kyoma ) for providing the mainland Chinapublic-interest speed test service