ComputeBox Provider Overview#

ComputeBox is a relatively new provider based in Mölln, Germany (near Hamburg). According to their own introduction, they run a small, independent datacenter powered by 100% renewable energy. They advertise bare-metal AMD EPYC Turin 9655, DDR5 RAM, and KIOXIA NVMe.

Based on publicly available information, their datacenter follows N+1 Tier 3 standards, has multiple uplinks to major carriers (including Telekom, Vodafone, CoreBackbone, Arelion, and Hurricane Electric), and is certified under ISO 27001:2017. The facility guarantees 99.98% uptime and has backup power for more than a week. Overall it looks quite solid and professional. Datacenter introduction.

The control panel is a newly self-developed one, but it looks very similar to the panel used by v6node. It is still under active development, with VNC and firewall features coming soon.

Current benchmarks look great, but given the “ghost ship” nature of these ultra-cheap offers, there may be overselling risks down the road. Buy with caution.

Test IP : 91.108.80.108

Official site : ComputeBox.de

Aerial view of the company building in a green environment

Plan Details#

Provider : ComputeBox
Type/Plan : Launch Promo
Processor : AMD EPYC 9655 96-Core Processor
Num of Core : 3 Cores
Memory : 6 GB
Storage : 5****0 GB NVMe
Bandwidth : 2****TB @ 10 Gbps IN | 10 Gbps OUT
Location : DE
Price : $4.45

Control Panel#

Compared to some older panels, this one is dramatically faster both in responsiveness and in how quickly it processes actions. From placing the order to the VM actually booting takes less than 10 seconds.

Signup doesn’t require ID verification, and there’s no MaxMind risk check. You just receive a verification code by email, fill in some basic info, and you’re done. Only several European countries and the US are available to choose from; Asian countries appear to be missing for now, but the provider says you can open a ticket to correct this. If you choose a European country, VAT will be included in the price.

Interface for selecting CPU and plan configuration options

Server status monitoring and detailed hardware specs

CPU plan promo showing special offer at €4.45 per month

Benchmarks#

When logging in with the Debian 12 template, you’ll find that root login is disabled and you need to escalate privileges from the sub-user.

Please log in as user computebox instead of root

CPU#

1
Architecture:             x86_64
2
  CPU op-mode(s):         32-bit, 64-bit
3
  Address sizes:          52 bits physical, 57 bits virtual
4
  Byte Order:             Little Endian
5
CPU(s):                   3
6
  On-line CPU(s) list:    0-2
7
Vendor ID:                AuthenticAMD
8
  Model name:             AMD EPYC 9655 96-Core Processor
9
    CPU family:           26
10
    Model:                2
11
    Thread(s) per core:   1
12
    Core(s) per socket:   3
13
    Socket(s):            1
14
    Stepping:             1
15
    BogoMIPS:             5199.99
16
    Flags:                fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm rep_good nopl
17
                          cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lah                          f_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core invpcid_single ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adju                          st bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves                           avx_vnni avx512_bf16 clzero xsaveerptr wbnoinvd arat npt lbrv nrip_save tsc_scale vmcb_clean flushbyasid pausefilter pfthreshold v_vmsave_vmload vgif avx512vbm                          i umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid bus_lock_detect movdiri movdir64b fsrm avx512_vp2inters                          ect flush_l1d arch_capabilities
18
Virtualization features:
19
  Virtualization:         AMD-V
20
  Hypervisor vendor:      KVM
21
  Virtualization type:    full
22
Caches (sum of all):
23
  L1d:                    192 KiB (3 instances)
24
  L1i:                    192 KiB (3 instances)
25
  L2:                     1.5 MiB (3 instances)
26
  L3:                     48 MiB (3 instances)
27
NUMA:
28
  NUMA node(s):           1
29
  NUMA node0 CPU(s):      0-2
30
Vulnerabilities:
31
  Gather data sampling:   Not affected
32
  Itlb multihit:          Not affected
33
  L1tf:                   Not affected
34
  Mds:                    Not affected
35
  Meltdown:               Not affected
36
  Mmio stale data:        Not affected
37
  Reg file data sampling: Not affected
38
  Retbleed:               Not affected
39
  Spec rstack overflow:   Not affected
40
  Spec store bypass:      Mitigation; Speculative Store Bypass disabled via prctl
41
  Spectre v1:             Mitigation; usercopy/swapgs barriers and __user pointer sanitization
42
  Spectre v2:             Mitigation; Enhanced / Automatic IBRS; IBPB conditional; STIBP disabled; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
43
  Srbds:                  Not affected
44
  Tsx async abort:        Not affected

Yabs#

1
Basic System Information:
2
---------------------------------
3
Uptime     : 0 days, 0 hours, 14 minutes
4
Processor  : AMD EPYC 9655 96-Core Processor
5
CPU cores  : 3 @ 2599.998 MHz
6
AES-NI     : ✔ Enabled
7
VM-x/AMD-V : ✔ Enabled
8
RAM        : 5.8 GiB
9
Swap       : 0.0 KiB
10
Disk       : 49.1 GiB
11
Distro     : Debian GNU/Linux 12 (bookworm)
12
Kernel     : 6.1.0-32-cloud-amd64
13
VM Type    : KVM
14
IPv4/IPv6  : ✔ Online / ❌ Offline
15

16
IPv4 Network Information:
17
---------------------------------
18
ISP        : Stephan Rakowski, trading as CSN Solutions e.K.
19
ASN        : AS212341 Stephan Rakowski, trading as CSN Solutions e.K.
20
Host       : Moritz Moeller Workstations
21
Location   : Mölln, Schleswig-Holstein (SH)
22
Country    : Germany
23

24
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda1):
25
---------------------------------
26
Block Size | 4k            (IOPS) | 64k           (IOPS)
27
  ------ | --- ---- | ---- ----
28
Read       | 681.32 MB/s (170.3k) | 5.02 GB/s    (78.5k)
29
Write      | 683.12 MB/s (170.7k) | 5.05 GB/s    (78.9k)
30
Total      | 1.36 GB/s   (341.1k) | 10.08 GB/s  (157.5k)
31
           |                      |
32
Block Size | 512k          (IOPS) | 1m            (IOPS)
33
  ------ | --- ---- | ---- ----
34
Read       | 5.92 GB/s    (11.5k) | 6.11 GB/s     (5.9k)
35
Write      | 6.23 GB/s    (12.1k) | 6.52 GB/s     (6.3k)
36
Total      | 12.15 GB/s   (23.7k) | 12.63 GB/s   (12.3k)
37

38
iperf3 Network Speed Tests (IPv4):
39
---------------------------------
40
Provider        | Location (Link)           | Send Speed      | Recv Speed      | Ping
41
----- | ----- | ---- | ---- | ----
42
Clouvider       | London, UK (10G)          | 4.71 Gbits/sec  | 714 Mbits/sec   | 14.8 ms
43
Eranium         | Amsterdam, NL (100G)      | 4.86 Gbits/sec  | 9.30 Gbits/sec  | 15.7 ms
44
Uztelecom       | Tashkent, UZ (10G)        | 2.09 Gbits/sec  | 2.01 Gbits/sec  | 94.6 ms
45
Leaseweb        | Singapore, SG (10G)       | 961 Mbits/sec   | 1.03 Gbits/sec  | 171 ms
46
Clouvider       | Los Angeles, CA, US (10G) | 871 Mbits/sec   | 1.17 Gbits/sec  | 155 ms
47
Leaseweb        | NYC, NY, US (10G)         | 1.77 Gbits/sec  | 2.21 Gbits/sec  | 84.6 ms
48
Edgoo           | Sao Paulo, BR (1G)        | 716 Mbits/sec   | 926 Mbits/sec   | 191 ms
49

50
Geekbench 6 BenchMark Test:
51
---------------------------------
52
Test            | Value
53
                |
54
Single Core     | 2897
55
Multi Core      | 7539
56
Full Test       | https://browser.geekbench.com/v6/cpu/11262024

GeekBench5#

1
Geekbench 5 测试结果
2

3
系统信息
4
  Operating System              Debian GNU/Linux 12 (bookworm)
5
  Kernel                        Linux 6.1.0-32-cloud-amd64 x86_64
6
  Model                         QEMU Standard PC (Q35 + ICH9, 2009)
7
  Motherboard                   N/A
8
  BIOS                          EFI Development Kit II / OVMF 3.20230228-4
9

10
处理器信息
11
  Name                          AMD EPYC 9655 96-Core Processor
12
  Topology                      1 Processor, 3 Cores
13
  Identifier                    AuthenticAMD Family 26 Model 2 Stepping 1
14
  Base Frequency                2.60 GHz
15
  L1 Instruction Cache          64.0 KB x 3
16
  L1 Data Cache                 64.0 KB x 3
17
  L2 Cache                      512 KB x 3
18
  L3 Cache                      16.0 MB
19

20
内存信息
21
  Size                          5.79 GB
22

23
单核测试分数：2197
24
多核测试分数：5887
25
详细结果链接：https://browser.geekbench.com/v5/cpu/23436432
26
可供参考链接：https://browser.geekbench.com/search?k=v5_cpu&q=AMD%20EPYC%209655

UnixBench#

1
========================================================================
2
   BYTE UNIX BenchMarks (Version 5.1.3)
3

4
   System: catcat: GNU/Linux
5
   OS: GNU/Linux -- 6.1.0-32-cloud-amd64 -- #1 SMP PREEMPT_DYNAMIC Debian 6.1.129-1 (2025-03-06)
6
   Machine: x86_64 (unknown)
7
   Language: en_US.utf8 (charmap="ANSI_X3.4-1968", collate="ANSI_X3.4-1968")
8
   CPU 0: AMD EPYC 9655 96-Core Processor (5200.0 bogomips)
9
          Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET
10
   CPU 1: AMD EPYC 9655 96-Core Processor (5200.0 bogomips)
11
          Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET
12
   CPU 2: AMD EPYC 9655 96-Core Processor (5200.0 bogomips)
13
          Hyper-Threading, x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, AMD virtualization, SYSCALL/SYSRET
14
   06:33:11 up 52 min,  3 users,  load average: 0.11, 0.07, 0.18; runlevel Mar
15

16
------------------------------------------------------------------------
17
BenchMark Run: Sat Mar 29 2025 06:33:11 - 07:01:06
18
3 CPUs in system; running 1 parallel copy of tests
19

20
Dhrystone 2 using register variables       87834313.3 lps   (10.0 s, 7 samples)
21
Double-Precision Whetstone                    11092.3 MWIPS (10.0 s, 7 samples)
22
Execl Throughput                               8089.9 lps   (30.0 s, 2 samples)
23
File Copy 1024 bufsize 2000 maxblocks       2911834.4 KBps  (30.0 s, 2 samples)
24
File Copy 256 bufsize 500 maxblocks          749698.8 KBps  (30.0 s, 2 samples)
25
File Copy 4096 bufsize 8000 maxblocks       9595443.9 KBps  (30.0 s, 2 samples)
26
Pipe Throughput                             4189689.2 lps   (10.0 s, 7 samples)
27
Pipe-based Context Switching                  76014.9 lps   (10.0 s, 7 samples)
28
Process Creation                              13898.5 lps   (30.0 s, 2 samples)
29
Shell Scripts (1 concurrent)                  22584.7 lpm   (60.0 s, 2 samples)
30
Shell Scripts (8 concurrent)                   5975.8 lpm   (60.0 s, 2 samples)
31
System Call Overhead                        3769905.8 lps   (10.0 s, 7 samples)
32

33
System BenchMarks Index Values               BASELINE       RESULT    INDEX
34
Dhrystone 2 using register variables         116700.0   87834313.3   7526.5
35
Double-Precision Whetstone                       55.0      11092.3   2016.8
36
Execl Throughput                                 43.0       8089.9   1881.4
37
File Copy 1024 bufsize 2000 maxblocks          3960.0    2911834.4   7353.1
38
File Copy 256 bufsize 500 maxblocks            1655.0     749698.8   4529.9
39
File Copy 4096 bufsize 8000 maxblocks          5800.0    9595443.9  16543.9
40
Pipe Throughput                               12440.0    4189689.2   3367.9
41
Pipe-based Context Switching                   4000.0      76014.9    190.0
42
Process Creation                                126.0      13898.5   1103.1
43
Shell Scripts (1 concurrent)                     42.4      22584.7   5326.6
44
Shell Scripts (8 concurrent)                      6.0       5975.8   9959.7
45
System Call Overhead                          15000.0    3769905.8   2513.3
46
                                                                   ========
47
System BenchMarks Index Score                                        3267.5
48

49
------------------------------------------------------------------------
50
BenchMark Run: Sat Mar 29 2025 07:01:06 - 07:29:01
51
3 CPUs in system; running 3 parallel copies of tests
52

53
Dhrystone 2 using register variables      262037231.4 lps   (10.0 s, 7 samples)
54
Double-Precision Whetstone                    33254.9 MWIPS (10.0 s, 7 samples)
55
Execl Throughput                              15084.9 lps   (30.0 s, 2 samples)
56
File Copy 1024 bufsize 2000 maxblocks       8419160.3 KBps  (30.0 s, 2 samples)
57
File Copy 256 bufsize 500 maxblocks         2230519.4 KBps  (30.0 s, 2 samples)
58
File Copy 4096 bufsize 8000 maxblocks      25156663.1 KBps  (30.0 s, 2 samples)
59
Pipe Throughput                            12503136.8 lps   (10.0 s, 7 samples)
60
Pipe-based Context Switching                1251634.9 lps   (10.0 s, 7 samples)
61
Process Creation                              45958.7 lps   (30.0 s, 2 samples)
62
Shell Scripts (1 concurrent)                  45277.7 lpm   (60.0 s, 2 samples)
63
Shell Scripts (8 concurrent)                   6502.3 lpm   (60.0 s, 2 samples)
64
System Call Overhead                       11311022.7 lps   (10.0 s, 7 samples)
65

66
System BenchMarks Index Values               BASELINE       RESULT    INDEX
67
Dhrystone 2 using register variables         116700.0  262037231.4  22453.9
68
Double-Precision Whetstone                       55.0      33254.9   6046.3
69
Execl Throughput                                 43.0      15084.9   3508.1
70
File Copy 1024 bufsize 2000 maxblocks          3960.0    8419160.3  21260.5
71
File Copy 256 bufsize 500 maxblocks            1655.0    2230519.4  13477.5
72
File Copy 4096 bufsize 8000 maxblocks          5800.0   25156663.1  43373.6
73
Pipe Throughput                               12440.0   12503136.8  10050.8
74
Pipe-based Context Switching                   4000.0    1251634.9   3129.1
75
Process Creation                                126.0      45958.7   3647.5
76
Shell Scripts (1 concurrent)                     42.4      45277.7  10678.7
77
Shell Scripts (8 concurrent)                      6.0       6502.3  10837.2
78
System Call Overhead                          15000.0   11311022.7   7540.7
79
                                                                   ========
80
System BenchMarks Index Score                                        9573.3

PerformanceTest Linux Benchmark#

1
AMD EPYC 9655 96-Core Processor (x86_64)
2
3 cores @ 0 MHz  |  5.8 GiB RAM
3
Number of Processes: 3  |  Test Iterations: 1  |  Test Duration: Medium
4
--------------------------------------------------------------------------
5
CPU Mark:                          10350
6
  Integer Math                     23019 Million Operations/s
7
  Floating Point Math              23021 Million Operations/s
8
  Prime Numbers                    90.5 Million Primes/s
9
  Sorting                          13578 Thousand Strings/s
10
  Encryption                       4797 MB/s
11
  Compression                      101920 KB/s
12
  CPU Single Threaded              3829 Million Operations/s
13
  Physics                          1514 Frames/s
14
  Extended Instructions (SSE)      8819 Million Matrices/s
15

16
Memory Mark:                       2453
17
  Database Operations              4511 Thousand Operations/s
18
  Memory Read Cached               34172 MB/s
19
  Memory Read Uncached             31785 MB/s
20
  Memory Write                     30466 MB/s
21
  Available RAM                    5476 Megabytes
22
  Memory Latency                   77 Nanoseconds
23
  Memory Threaded                  94907 MB/s
24
--------------------------------------------------------------------------

WordPress BenchMark#

Server performance test results, all items rated 100% excellent

Server performance test result with an overall score of 9.9

IP Quality#

IP risk analysis report with multiple security indicators

Network Quality#

Return Path Routing#

Network trace report showing latency on multiple paths

Traceroute output displaying multiple international hops

Traceroute results showing multiple hop details

BGP Info#

Network topology diagram showing connections to multiple ISPs

Side Note#

Starting next week, HostBrr will also launch promo configurations for 5th-gen AMD-based VPS/VDS. The current pricing from ComputeBox might even push HostBrr to adjust their prices—who knows. Once HostBrr goes live with their offer, I’ll grab a box and run benchmarks as well.

ComputeBox Germany AMD EPYC 9655 Benchmark