前言#

随着黑五的来到，沉寂了半年的Hostbrr也推出了新的促销,分别有AMD Threadripper 9995WX Pro ，AMD Epyc 9655 ,7欧存储盒以及1T1欧的块存储。HostBrr是德国商家，主打高性价和低价存储。机器谈不上非常好，但是胜在价格便宜。支持比较慢，就老板一个人。机器大多来自DataForest的杜甫，更新换代硬件很勤快。存储盒7欧一年500G拿来当异地备份很nice。

对比来看，还是不如9950x，但是核心多，可以避免9950x超售严重的情况（虽然都是凭良心） CPU是9950x>9995wx>9655，请根据自己的需求购买套餐。机器网络存在问题，大流量测试会直接断流。我猜测应该是机房ddos防火墙策略导致的，LowEndTalk上很多人也反馈，有需要的小伙伴谨慎购买。

其他测评请见：https://catcat.blog/archive.html?tag=HostBrr

套餐	CPU	内存	硬盘	流量	价格	AFF 查看链接
Threadripper 1 GB	1* 9995WX Pro	1 GB DDR5 ECC	15 GB NVMe Gen4	5000 GB @ 10 Gbps	15 €/年	查看
Threadripper 3 GB	1* 9995WX Pro	3 GB DDR5 ECC	45 GB NVMe Gen4	10000 GB @ 10 Gbps	25 €/年	查看
Threadripper 6 GB	2* 9995WX Pro	6 GB DDR5 ECC	70 GB NVMe Gen4	15000 GB @ 10 Gbps	5 €/月；50 €/年	查看
Threadripper 12 GB	3* 9995WX Pro	12 GB DDR5 ECC	150 GB NVMe Gen4	15000 GB @ 10 Gbps	10 €/月；100 €/年	查看
MINI-EPYC BF2025	1* EPYC 9655	4 GB DDR5 ECC	40/80 GB NVMe Gen4	10/20 TB @ 10 Gbps	3 €/月；30 €/年	查看
NICE-EPYC BF2025	2* EPYC 9655	6 GB DDR5 ECC	60/120 GB NVMe Gen4	10/20 TB @ 10 Gbps	5 €/月；50 €/年	查看
EPIC-EPYC BF2025	4* EPYC 9655	16 GB DDR5 ECC	160/320 GB NVMe Gen4	40/80 TB @ 10 Gbps	9 €/月；90 €/年	查看
StorageBox 500 GB	—	—	500 GB RAID-60 + NVMe 缓存	2.5 TB	7 €/年	查看
StorageBox 1 TB	—	—	1 TB RAID-60 + NVMe 缓存	5 TB	2.5 €/月；15 €/年（BFBOX）	查看
StorageBox 2 TB	—	—	2 TB RAID-60 + NVMe 缓存	10 TB	4 €/月；27 €/年（BFBOX）	查看
StorageBox 4 TB	—	—	4 TB RAID-60 + NVMe 缓存	20 TB	8.5 €/月；51 €/年（BFBOX）	查看

测评#

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):                      2
6
  On-line CPU(s) list:       0,1
7
Vendor ID:                   AuthenticAMD
8
  Model name:                AMD Ryzen Threadripper PRO 9995WX 96-Cores
9
    CPU family:              26
10
    Model:                   8
11
    Thread(s) per core:      1
12
    Core(s) per socket:      1
13
    Socket(s):               2
14
    Stepping:                1
15
    BogoMIPS:                5000.03
16
    Flags:                   fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 syscall nx mmxext fxsr_opt pd
17
                             pe1gb rdtscp lm rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe pop
18
                             cnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefet
19
                             ch osvw perfctr_core ssbd ibrs ibpb stibp vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid avx512f avx512dq rdseed
20
                             adx smap avx512ifma clflushopt clwb avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves avx_vnni avx512_bf16 clzer
21
                             o xsaveerptr wbnoinvd arat npt lbrv nrip_save tsc_scale vmcb_clean pausefilter pfthreshold v_vmsave_vmload vgif avx512vbmi umip
22
                              pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq la57 rdpid movdiri movdir64b fsrm avx51
23
                             2_vp2intersect arch_capabilities
24
Virtualization features:
25
  Virtualization:            AMD-V
26
  Hypervisor vendor:         KVM
27
  Virtualization type:       full
28
Caches (sum of all):
29
  L1d:                       96 KiB (2 instances)
30
  L1i:                       64 KiB (2 instances)
31
  L2:                        2 MiB (2 instances)
32
  L3:                        768 MiB (2 instances)
33
NUMA:
34
  NUMA node(s):              1
35
  NUMA node0 CPU(s):         0,1
36
Vulnerabilities:
37
  Gather data sampling:      Not affected
38
  Indirect target selection: Not affected
39
  Itlb multihit:             Not affected
40
  L1tf:                      Not affected
41
  Mds:                       Not affected
42
  Meltdown:                  Not affected
43
  Mmio stale data:           Not affected
44
  Reg file data sampling:    Not affected
45
  Retbleed:                  Not affected
46
  Spec rstack overflow:      Not affected
47
  Spec store bypass:         Mitigation; Speculative Store Bypass disabled via prctl
48
  Spectre v1:                Mitigation; usercopy/swapgs barriers and __user pointer sanitization
49
  Spectre v2:                Mitigation; Retpolines; IBPB conditional; IBRS_FW; STIBP disabled; RSB filling; PBRSB-eIBRS Not affected; BHI Not affected
50
  Srbds:                     Not affected
51
  Tsa:                       Not affected

Yabs#

1
Basic System Information:
2
---------------------------------
3
Uptime     : 0 days, 0 hours, 5 minutes
4
Processor  : AMD Ryzen Threadripper PRO 9995WX 96-Cores
5
CPU cores  : 2 @ 2500.018 MHz
6
AES-NI     : ✔ Enabled
7
VM-x/AMD-V : ✔ Enabled
8
RAM        : 5.8 GiB
9
Swap       : 0.0 KiB
10
Disk       : 68.9 GiB
11
Distro     : Debian GNU/Linux 13 (trixie)
12
Kernel     : 6.12.38+deb13-amd64
13
VM Type    : KVM
14
IPv4/IPv6  : ✔ Online / ✔ Online
15

16
IPv6 Network Information:
17
---------------------------------
18
ISP        : IPPv
19
ASN        : AS48314 Michael Sebastian Schinzel trading as IP-Projects GmbH \u0026 Co. KG
20
Host       : IP-Projects GmbH \u0026 Co.
21
Location   : Waldbrunn, Bavaria (BY)
22
Country    : Germany
23

24
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vda3):
25
---------------------------------
26
Block Size | 4k            (IOPS) | 64k           (IOPS)
27
  ------   | ---            ----  | ----           ----
28
Read       | 435.69 MB/s (108.9k) | 2.08 GB/s    (32.5k)
29
Write      | 436.84 MB/s (109.2k) | 2.09 GB/s    (32.7k)
30
Total      | 872.54 MB/s (218.1k) | 4.18 GB/s    (65.3k)
31
           |                      |
32
Block Size | 512k          (IOPS) | 1m            (IOPS)
33
  ------   | ---            ----  | ----           ----
34
Read       | 1.03 GB/s     (2.0k) | 958.24 MB/s    (935)
35
Write      | 1.08 GB/s     (2.1k) | 1.02 GB/s      (998)
36
Total      | 2.12 GB/s     (4.1k) | 1.98 GB/s     (1.9k)
37

38

39
fio Disk Speed Tests (Mixed R/W 50/50) (Partition /dev/vdb):
40
---------------------------------
41
Block Size | 4k            (IOPS) | 64k           (IOPS)
42
  ------   | ---            ----  | ----           ----
43
Read       | 1.61 MB/s      (403) | 15.66 MB/s     (244)
44
Write      | 1.64 MB/s      (410) | 16.19 MB/s     (253)
45
Total      | 3.25 MB/s      (813) | 31.86 MB/s     (497)
46
           |                      |
47
Block Size | 512k          (IOPS) | 1m            (IOPS)
48
  ------   | ---            ----  | ----           ----
49
Read       | 48.14 MB/s      (94) | 57.95 MB/s      (56)
50
Write      | 51.04 MB/s      (99) | 62.58 MB/s      (61)
51
Total      | 99.19 MB/s     (193) | 120.53 MB/s    (117)
52

53

54
Geekbench 5 Benchmark Test:
55
---------------------------------
56
Test            | Value
57
                |
58
Single Core     | 2035
59
Multi Core      | 3807
60
Full Test       | https://browser.geekbench.com/v5/cpu/23929460
61

62
Geekbench 6 Benchmark Test:
63
---------------------------------
64
Test            | Value
65
                |
66
Single Core     | 2733
67
Multi Core      | 4919
68
Full Test       | https://browser.geekbench.com/v6/cpu/15257887

UnixBench#

1
------------------------------------------------------------------------
2
Benchmark Run: Thu Nov 27 2025 16:17:58 - 16:46:10
3
2 CPUs in system; running 1 parallel copy of tests
4

5
Dhrystone 2 using register variables       90992987.4 lps   (10.0 s, 7 samples)
6
Double-Precision Whetstone                    12674.0 MWIPS (9.9 s, 7 samples)
7
Execl Throughput                               6704.0 lps   (29.9 s, 2 samples)
8
File Copy 1024 bufsize 2000 maxblocks       2981876.9 KBps  (30.0 s, 2 samples)
9
File Copy 256 bufsize 500 maxblocks          789597.8 KBps  (30.0 s, 2 samples)
10
File Copy 4096 bufsize 8000 maxblocks       9007488.3 KBps  (30.0 s, 2 samples)
11
Pipe Throughput                             4790714.9 lps   (10.0 s, 7 samples)
12
Pipe-based Context Switching                 110712.7 lps   (10.0 s, 7 samples)
13
Process Creation                              15583.0 lps   (30.0 s, 2 samples)
14
Shell Scripts (1 concurrent)                  21671.8 lpm   (60.0 s, 2 samples)
15
Shell Scripts (8 concurrent)                   4525.7 lpm   (60.0 s, 2 samples)
16
System Call Overhead                        4034388.2 lps   (10.0 s, 7 samples)
17

18
System Benchmarks Index Values               BASELINE       RESULT    INDEX
19
Dhrystone 2 using register variables         116700.0   90992987.4   7797.2
20
Double-Precision Whetstone                       55.0      12674.0   2304.4
21
Execl Throughput                                 43.0       6704.0   1559.1
22
File Copy 1024 bufsize 2000 maxblocks          3960.0    2981876.9   7530.0
23
File Copy 256 bufsize 500 maxblocks            1655.0     789597.8   4771.0
24
File Copy 4096 bufsize 8000 maxblocks          5800.0    9007488.3  15530.2
25
Pipe Throughput                               12440.0    4790714.9   3851.1
26
Pipe-based Context Switching                   4000.0     110712.7    276.8
27
Process Creation                                126.0      15583.0   1236.7
28
Shell Scripts (1 concurrent)                     42.4      21671.8   5111.3
29
Shell Scripts (8 concurrent)                      6.0       4525.7   7542.8
30
System Call Overhead                          15000.0    4034388.2   2689.6
31
                                                                   ========
32
System Benchmarks Index Score                                        3368.8
33

34
------------------------------------------------------------------------
35
Benchmark Run: Thu Nov 27 2025 16:46:10 - 17:14:24
36
2 CPUs in system; running 2 parallel copies of tests
37

38
Dhrystone 2 using register variables      180975812.8 lps   (10.0 s, 7 samples)
39
Double-Precision Whetstone                    25087.6 MWIPS (9.9 s, 7 samples)
40
Execl Throughput                              10540.6 lps   (30.0 s, 2 samples)
41
File Copy 1024 bufsize 2000 maxblocks       5961936.2 KBps  (30.0 s, 2 samples)
42
File Copy 256 bufsize 500 maxblocks         1566480.4 KBps  (30.0 s, 2 samples)
43
File Copy 4096 bufsize 8000 maxblocks      16500156.0 KBps  (30.0 s, 2 samples)
44
Pipe Throughput                             9570891.9 lps   (10.0 s, 7 samples)
45
Pipe-based Context Switching                1075862.4 lps   (10.0 s, 7 samples)
46
Process Creation                              39264.9 lps   (30.0 s, 2 samples)
47
Shell Scripts (1 concurrent)                  31181.8 lpm   (60.0 s, 2 samples)
48
Shell Scripts (8 concurrent)                   4060.1 lpm   (60.0 s, 2 samples)
49
System Call Overhead                        7978136.7 lps   (10.0 s, 7 samples)
50

51
System Benchmarks Index Values               BASELINE       RESULT    INDEX
52
Dhrystone 2 using register variables         116700.0  180975812.8  15507.8
53
Double-Precision Whetstone                       55.0      25087.6   4561.4
54
Execl Throughput                                 43.0      10540.6   2451.3
55
File Copy 1024 bufsize 2000 maxblocks          3960.0    5961936.2  15055.4
56
File Copy 256 bufsize 500 maxblocks            1655.0    1566480.4   9465.1
57
File Copy 4096 bufsize 8000 maxblocks          5800.0   16500156.0  28448.5
58
Pipe Throughput                               12440.0    9570891.9   7693.6
59
Pipe-based Context Switching                   4000.0    1075862.4   2689.7
60
Process Creation                                126.0      39264.9   3116.3
61
Shell Scripts (1 concurrent)                     42.4      31181.8   7354.2
62
Shell Scripts (8 concurrent)                      6.0       4060.1   6766.8
63
System Call Overhead                          15000.0    7978136.7   5318.8
64
                                                                   ========
65
System Benchmarks Index Score                                        6917.1

PassMark PerformanceTest Linux#

1
AMD Ryzen Threadripper PRO 9995WX 96-Cores (x86_64)
2
2 cores @ 0 MHz  |  5.8 GiB RAM
3
Number of Processes: 2  |  Test Iterations: 1  |  Test Duration: Medium
4
--------------------------------------------------------------------------------
5
CPU Mark:                          7053
6
  Integer Math                     15659 Million Operations/s
7
  Floating Point Math              15722 Million Operations/s
8
  Prime Numbers                    56.4 Million Primes/s
9
  Sorting                          9420 Thousand Strings/s
10
  Encryption                       3276 MB/s
11
  Compression                      67786 KB/s
12
  CPU Single Threaded              3798 Million Operations/s
13
  Physics                          993 Frames/s
14
  Extended Instructions (SSE)      5687 Million Matrices/s
15

16
Memory Mark:                       2030
17
  Database Operations              2612 Thousand Operations/s
18
  Memory Read Cached               34971 MB/s
19
  Memory Read Uncached             29393 MB/s
20
  Memory Write                     23505 MB/s
21
  Available RAM                    4153 Megabytes
22
  Memory Latency                   89 Nanoseconds
23
  Memory Threaded                  41157 MB/s
24
--------------------------------------------------------------------------------

Hostbrr AMD Ryzen Threadripper PRO 9995WX 测评

前言#

测评#

CPU#

Yabs#

UnixBench#

PassMark PerformanceTest Linux#

IP测试#

读写50G测试#

编译本站速度#