Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

VPS Benchmarks7~months ago~Published LetsHosting
0
Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

In an era where Asia-Pacific businesses demand low-latency infrastructure to support expanding digital footprints, Melbicom’s KVM-8-TYO Cloud Server emerges as a compelling option for enterprise IT teams. This test evaluates a shared vCPU VPS plan hosted in Tokyo, Japan, powered by an AMD EPYC 7702P 64-core processor with 8 allocated cores, 16 GB RAM, 160 GB NVMe SSD storage, and up to 1 Gbps bandwidth (throttled to 100 Mbps beyond a 5 TB monthly quota).

Benchmarks reveal robust compute and storage: Sysbench scores hit 1,627 (single-thread) and 12,915 (multi-thread), with Geekbench at 1,241/6,377 single/multi-core. Memory blazes at 42,768 MB/s read, 19,578 MB/s write; NVMe delivers 249 MB/s (4K, 62.3k IOPS) to 4.41 GB/s (1M, 4.3k IOPS).

Network peaks at 946 Mbps upload but dips to 74 Mbps receive in the US; pings range 109-336 ms. Streaming partially unlocks (YouTube, Amazon Prime), but Netflix/Disney+ falter. IP quality scores 56/100 trust, with low abuse but a 65 fraud score.

The VPS excels for regional workloads—like a sprinter on a short track—perfect for Tokyo-centric apps. Network lag and IP reputation may hinder global or compliance-heavy use. It’s a cost-effective pick for APAC-focused IT teams.

About Melbicom

Melbicom.net stands out as a premier hosting provider, delivering robust solutions tailored for businesses, developers, and high-traffic websites worldwide. Established in 2005, Melbicom has evolved from a niche bandwidth specialist into a full-spectrum hosting powerhouse, headquartered in New York with a global footprint. Committed to innovation and reliability, the company leverages cutting-edge technology to ensure seamless connectivity and optimal performance, serving over 10,000 clients across various industries, including e-commerce, gaming, and enterprise IT.

At its core, Melbicom.net offers an impressive array of services designed to scale with your needs. Virtual Private Servers (VPS) start with flexible plans powered by KVM virtualization, featuring SSD storage, unlimited bandwidth, and root access for customization. For demanding applications, dedicated servers provide bare-metal power with Intel Xeon processors, up to 1TB RAM, and RAID-configured NVMe drives—ideal for databases or media streaming.

Colocation services excel in their state-of-the-art facilities, offering secure racks, 24/7 power redundancy, and direct carrier access in key hubs such as New York, Amsterdam, and Hong Kong. What sets Melbicom apart is its IP transit prowess: as a Tier-1 provider, it boasts peering with over 500 networks, delivering low-latency global reach at speeds up to 100Gbps. DDoS protection is baked in, with unlimited mitigation to safeguard against threats.

Pricing

Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

Official Website: https://www.Melbicom.net

Coupon

The merchant did not provide

Basic Information

 Processor         : AMD EPYC 7702P 64-Core Processor
 CPU Numbers       : 8
 CPU Frequency     : 1996.250 MHz
 CPU Cache         : L1: 256.00 KB / L2: 4.00 MB / L3: 2048.00 MB
 AES-NI            : ✔ Enabled
 VM-x/AMD-V        : ✔ Enabled
 RAM               : 456.01 MiB / 15.63 GiB
 Swap              : 0 KiB / 1024.00 MiB
 Disk Space        : 5.34 GiB / 157.39 GiB
 Boot Disk         : /dev/vda1
 Uptime            : 0 days, 12 hour 4 min
 Loads             : 0.43, 0.15, 0.04
 OS Release        : Ubuntu 22.04.1 LTS (x86_64)
 Arch              : x86_64 (64 Bit)
 Kernel Version    : 5.15.0-46-generic
 TCP Acceleration  : bbr
 VM Type           : Microsoft Hyper-V
 NAT Type          : Full Cone
 IPV4 ASN          : AS8849 Melbikomas UAB
 IPV4 Location     : Tokyo / Tokyo / JP

CPU Benchmark

---------------------------CPU-Sysbench-Test----------------------------
 -> CPU test in progress (Fast Mode, 1-Pass @ 5sec)
 1 Thread(s) Test:              1627 Scores
 8 Thread(s) Test:              12915 Scores
Geekbench 6 Benchmark Test:
---------------------------------
Test            | Value                         
                |                               
Single Core     | 1241                          
Multi Core      | 6377                          
Full Test       | https://browser.geekbench.com/v6/cpu/14444679

YABS completed in 6 min 29 sec

RAM Read/Write Tests

----------------------------Memory-Test---------------------------------
 -> Memory Test (Fast Mode, 1-Pass @ 5sec)
 Single Read Test:              42768.39 MB/s
 Single Write Test:             19578.80 MB/s
Running memory speed test with the following options:
  block size: 1KiB
  total size: 102400MiB
  operation: read
  scope: global

Initializing worker threads...

Threads started!

Total operations: 64626887 (6461755.11 per second)

63112.19 MiB transferred (6310.31 MiB/sec)


General statistics:
    total time:                          10.0001s
    total number of events:              64626887

Latency (ms):
         min:                                    0.00
         avg:                                    0.00
         max:                                    1.77
         95th percentile:                        0.00
         sum:                                 3198.99

Threads fairness:
    events (avg/stddev):           64626887.0000/0.00
    execution time (avg/stddev):   3.1990/0.00
Running memory speed test with the following options:
  block size: 1KiB
  total size: 102400MiB
  operation: write
  scope: global

Initializing worker threads...

Threads started!

Total operations: 54447287 (5444156.99 per second)

53171.18 MiB transferred (5316.56 MiB/sec)


General statistics:
    total time:                          10.0001s
    total number of events:              54447287

Latency (ms):
         min:                                    0.00
         avg:                                    0.00
         max:                                    2.65
         95th percentile:                        0.00
         sum:                                 4227.13

Threads fairness:
    events (avg/stddev):           54447287.0000/0.00
    execution time (avg/stddev):   4.2271/0.00

Disk Read/Write Tests

------------------------Disk-dd-Read/Write-Test-------------------------
 -> Disk test in progress (4K Block/1M Block, Direct Mode)
 Block Size             Write Test                              Read Test
 100MB-4K Block         33.7 MB/s (8235 IOPS, 3.11s)            36.6 MB/s (8932 IOPS, 2.87s)
 1GB-1M Block           1.2 GB/s (1139 IOPS, 0.88s)             1.3 GB/s (1212 IOPS, 0.82s)
-----------------------Disk-fio-Read/Write-Test-------------------------
Block Size | 4k            (IOPS) | 64k           (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 124.59 MB/s  (31.1k) | 1.78 GB/s    (27.8k)
Write      | 124.92 MB/s  (31.2k) | 1.79 GB/s    (27.9k)
Total      | 249.52 MB/s  (62.3k) | 3.57 GB/s    (55.8k)
           |                      |                     
Block Size | 512k          (IOPS) | 1m            (IOPS)
  ------   | ---            ----  | ----           ---- 
Read       | 2.06 GB/s     (4.0k) | 2.13 GB/s     (2.0k)
Write      | 2.17 GB/s     (4.2k) | 2.27 GB/s     (2.2k)
Total      | 4.24 GB/s     (8.2k) | 4.41 GB/s     (4.3k)

Iperf3 Network Tests

iperf3 Network Speed Tests (IPv4):
---------------------------------
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping           
Clouvider       | London, UK (10G)          | 619 Mbits/sec   | 272 Mbits/sec   | 238 ms         
Eranium         | Amsterdam, NL (100G)      | 767 Mbits/sec   | 778 Mbits/sec   | 253 ms         
Uztelecom       | Tashkent, UZ (10G)        | 740 Mbits/sec   | 334 Mbits/sec   | 257 ms         
Leaseweb        | Singapore, SG (10G)       | 946 Mbits/sec   | 763 Mbits/sec   | --             
Clouvider       | Los Angeles, CA, US (10G) | 697 Mbits/sec   | 74.2 Mbits/sec  | 109 ms         
Leaseweb        | NYC, NY, US (10G)         | 847 Mbits/sec   | 772 Mbits/sec   | 165 ms         
Edgoo           | Sao Paulo, BR (1G)        | 639 Mbits/sec   | 336 Mbits/sec   | 336 ms   

UnixBench Tests

Benchmark Run: Tue Oct 14 2025 07:31:43 - 07:59:42
8 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables       38729086.9 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                     7331.4 MWIPS (10.0 s, 7 samples)
Execl Throughput                               3681.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1068580.1 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          291620.9 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2871961.9 KBps  (30.0 s, 2 samples)
Pipe Throughput                             1612105.8 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                  55796.3 lps   (10.0 s, 7 samples)
Process Creation                               7464.2 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  11223.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   4513.3 lpm   (60.0 s, 2 samples)
System Call Overhead                        1777899.4 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0   38729086.9   3318.7
Double-Precision Whetstone                       55.0       7331.4   1333.0
Execl Throughput                                 43.0       3681.3    856.1
File Copy 1024 bufsize 2000 maxblocks          3960.0    1068580.1   2698.4
File Copy 256 bufsize 500 maxblocks            1655.0     291620.9   1762.1
File Copy 4096 bufsize 8000 maxblocks          5800.0    2871961.9   4951.7
Pipe Throughput                               12440.0    1612105.8   1295.9
Pipe-based Context Switching                   4000.0      55796.3    139.5
Process Creation                                126.0       7464.2    592.4
Shell Scripts (1 concurrent)                     42.4      11223.2   2647.0
Shell Scripts (8 concurrent)                      6.0       4513.3   7522.1
System Call Overhead                          15000.0    1777899.4   1185.3
                                                                   ========
System Benchmarks Index Score                                        1570.6

------------------------------------------------------------------------
Benchmark Run: Tue Oct 14 2025 07:59:42 - 08:27:43
8 CPUs in system; running 8 parallel copies of tests

Dhrystone 2 using register variables      305656772.0 lps   (10.0 s, 7 samples)
Double-Precision Whetstone                    59193.8 MWIPS (9.9 s, 7 samples)
Execl Throughput                              16604.3 lps   (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks       1081607.4 KBps  (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks          320221.7 KBps  (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks       2785495.1 KBps  (30.0 s, 2 samples)
Pipe Throughput                            12955148.6 lps   (10.0 s, 7 samples)
Pipe-based Context Switching                 692378.4 lps   (10.0 s, 7 samples)
Process Creation                              32582.3 lps   (30.0 s, 2 samples)
Shell Scripts (1 concurrent)                  46669.2 lpm   (60.0 s, 2 samples)
Shell Scripts (8 concurrent)                   6607.3 lpm   (60.0 s, 2 samples)
System Call Overhead                        3638505.6 lps   (10.0 s, 7 samples)

System Benchmarks Index Values               BASELINE       RESULT    INDEX
Dhrystone 2 using register variables         116700.0  305656772.0  26191.7
Double-Precision Whetstone                       55.0      59193.8  10762.5
Execl Throughput                                 43.0      16604.3   3861.5
File Copy 1024 bufsize 2000 maxblocks          3960.0    1081607.4   2731.3
File Copy 256 bufsize 500 maxblocks            1655.0     320221.7   1934.9
File Copy 4096 bufsize 8000 maxblocks          5800.0    2785495.1   4802.6
Pipe Throughput                               12440.0   12955148.6  10414.1
Pipe-based Context Switching                   4000.0     692378.4   1730.9
Process Creation                                126.0      32582.3   2585.9
Shell Scripts (1 concurrent)                     42.4      46669.2  11006.9
Shell Scripts (8 concurrent)                      6.0       6607.3  11012.1
System Call Overhead                          15000.0    3638505.6   2425.7
                                                                   ========
System Benchmarks Index Score                                        5186.2

======= Script description and score comparison completed! ======= 

Streaming Unlock Tests

============[ Multination ]============
 Dazn:                                  Failed (Error: )
 Disney+:                               No (IP Banned By Disney+ 1)
 Netflix:                               Originals Only
 YouTube Premium:                       Yes (Region: JP)
 Amazon Prime Video:                    Yes (Region: LT)
 TVBAnywhere+:                          Yes
 Spotify Registration:                  No
 OneTrust Region:                       JP [Tokyo]
 iQyi Oversea Region:                   INTL
 Bing Region:                           JP (Risky)
 Apple Region:                          LT
 YouTube CDN:                           Tokyo
 Netflix Preferred CDN:                 Tokyo
 ChatGPT:                               Yes
 Google Gemini:                         Yes (Region: JPN)
 Claude:                                Yes
 Wikipedia Editability:                 Yes
 Google Play Store:                     Japan 
 Google Search CAPTCHA Free:            Yes
 Steam Currency:                        JPY
 ---Forum---
 Reddit:                                No
 ---Game---
 SD Gundam G Generation Eternal:        No
=======================================

IP Quality Detection

Data for reference only, does not represent 100% accurate, if and the actual situation is not consistent with the manual query multiple database comparison
The following is the number of each database, the output will come with the corresponding number of the database source
ipinfo  databases [0] | scamalytics databases [1] | virustotal  databases  [2] | abuseipdb databases   [3] | ip2location databases    [4]
ip-api  databases [5] | ipwhois databases     [6] | ipregistry  databases  [7] | ipdata databases      [8] | db-ip databases          [9]
ipapiis databases [A] | ipapicom databases    [B] | bigdatacloud databases [C] | dkly databases        [D] | ipqualityscore databases [E]
IPV4:
Security Score:
Reputation: 0 [2] 
TrustScore: 56 [8]
VpnScore: 3 [8] 
ProxyScore: 33 [8] 
CommunityVoteHarmless: 0 [2] 
CommunityVoteMalicious: 0 [2] 
ThreatScore: 97 [8] 
FraudScore: 65 [E] 
AbuseScore: 0 [3] 
ASNAbuseScore: 0.0025 (Low) [A] 
CompannyAbuseScore: 0 (Very Low) [A] 
ThreatLevel: low [9] 
Blacklist_Records_Statistics(how many blacklisted websites have records):
HarmlessnessRecords: 1 [2]  MaliciousRecords: 0 [2]  SuspiciousRecords: 0 [2]  NoRecords: 94 [2]  
Security Info:
UsageType: business [8] hosting [0 7 A] corporate [9] DataCenter/WebHosting/Transit [3] hosting ASN [C]
CompanyType: hosting [0 7 A] 
IsCloudProvider: Yes [7] 
IsDatacenter: Yes [0 A C] No [5 6 8]
IsMobile: Yes [E] No [5 A C]
IsProxy: No [0 4 6 7 8 9 A C] Yes [5 E]
IsVpn: No [0 6 C] Yes [7 A E]
IsTor: No [0 3 6 7 8 A C E] 
IsTorExit: No [7] 
IsCrawler: No [9 A E] 
IsAnonymous: Yes [7] No [6 8]
IsAttacker: No [7 8]
IsAbuser: No [7 8 A C E] 
IsThreat: No [7 8 C] 
IsRelay: No [0 7 8 C] 
IsBogon: No [7 8 A C] 
IsBot: No [E] 
DNS-BlackList: 315(Total_Check) 0(Clean) 5(Blacklisted) 310(Other) 
Google search feasibility: NO

Email Port Detection

Platform  SMTP  SMTPS POP3  POP3S IMAP  IMAPS
LocalPort ✔     ✔     ✔     ✔     ✔     ✔    
QQ        ✘     ✘     ✘     ✘     ✘     ✘    
163       ✘     ✘     ✔     ✘     ✔     ✘    
Sohu      ✘     ✘     ✔     ✘     ✔     ✘    
Yandex    ✘     ✘     ✔     ✘     ✔     ✘    
Gmail     ✘     ✘     ✘     ✘     ✘     ✘    
Outlook   ✘     ✘     ✔     ✘     ✔     ✘    
Office365 ✘     ✘     ✔     ✘     ✔     ✘    
Yahoo     ✘     ✘     ✘     ✘     ✘     ✘    
MailCOM   ✘     ✘     ✔     ✘     ✔     ✘    
MailRU    ✘     ✘     ✘     ✘     ✔     ✘    
AOL       ✘     ✘     ✘     ✘     ✘     ✘    
GMX       ✘     ✘     ✔     ✘     ✔     ✘    
Sina      ✘     ✘     ✔     ✘     ✔     ✘    
Apple     ✘     ✘     ✘     ✘     ✘     ✘    
FastMail  ✘     ✘     ✘     ✘     ✘     ✘    
ProtonMail✘     ✘     ✘     ✘     ✘     ✘    
MXRoute   ✘     ✘     ✔     ✘     ✔     ✘    
Namecrane ✘     ✘     ✔     ✘     ✔     ✘    
XYAMail   ✘     ✘     ✘     ✘     ✘     ✘    
ZohoMail  ✘     ✘     ✘     ✘     ✘     ✘    
Inbox_eu  ✘     ✘     ✔     ✘     ✘     ✘    
Free_fr   ✘     ✘     ✔     ✘     ✔     ✘   

Looking Glass

Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review
  • CityTokyo
  • CountryJapan
  • LocationAsia
  • Space
  • Electricity
  • Capacity
  • Network1-100 Gbps (per server)
  • CertificationsTier III

Test IPv4: 203.190.0.130

Download File: 1000MB

Global Latency Tests

Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

Query: ping 203.190.0.90, interval: 5s, dense mode: no

Probing 203.190.0.90 – RIPE NCC /22 AS8849 MELBICOM-EU-AS – Japan, Tokyo, Tokyo [hosting]

CloudFlare Internet Tests

Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

Admin Control Panel

Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review Melbicom Japan Tokyo AMD EPYC 7702P High-Performance KVM VPS Test and Review

Conclusion

Key metrics from the benchmarks paint a picture of solid, if regionally nuanced, capabilities. CPU performance shines in Sysbench tests, hitting 1,627 scores on a single thread and scaling impressively to 12,915 on eight threads, while Geekbench 6 clocks in at 1,241 single-core and 6,377 multi-core—adequate for parallel workloads. Memory throughput impresses with 42,768 MB/s reads and 19,578 MB/s writes, backed by low-latency figures under 2 ms. NVMe storage delivers brisk speeds: fio tests show 249 MB/s total for 4K blocks (62.3k IOPS) and up to 4.41 GB/s for 1M blocks (4.3k IOPS), ideal for database ops.

Network tests via iPerf3 reveal upload peaks of 946 Mbps to Singapore and 847 Mbps to New York, though receive lag at 74 Mbps to Los Angeles; pings hover around 109-336 ms globally. Streaming partially succeeds—YouTube Premium and Amazon Prime work in JP/LT regions, but Netflix limits to originals, and Disney+ blocks the IP. IP quality scores a 56/100 trust rating, with low abuse (0.0025) but a 65 fraud score, flagging minor proxy risks; email ports favor local but falter on Gmail/Yahoo.

Delving deeper, these results underscore Melbicom’s strengths in compute and storage, where the EPYC processor’s multi-threading muscle handles bursts like a well-tuned engine in a rally—efficient for dev/test cycles or e-commerce backends. Worth noting, the NVMe’s IOPS rival pricier rivals, suggesting it’s no slouch for I/O-intensive apps. Yet, network inconsistencies emerge as a soft spot: that 74 Mbps receive to the US feels like a sports car idling in traffic, potentially bottlenecking cross-Pacific data syncs for hybrid teams.

In wrapping up, the Melbicom KVM-8-TYO VPS delivers respectable value for Asia-centric deployments, blending EPYC-driven grunt with NVMe zip at an accessible price point—think reliable workhorse for mid-tier IT stacks. For administrators prioritizing Tokyo latency under 250 ms, it’s a smart pick, especially with unmetered transfers post-quota. That said, if your ops span continents or crave flawless streaming compliance, pair it with a CDN or scout alternatives for polished IP hygiene. Prospective buyers should trial it for workload-specific fits; in a crowded VPS market, this one’s edge lies in unpretentious power—grab it if APAC agility tops your list.

Official Website: https://www.Melbicom.net

Related Posts