iwinv High Performance Gen 2가 추가되어서 벤치를 돌려봤다.
2025.05.24 - [클라우드] - iwinv 인스턴스 타입 비교 벤치마크
iwinv 인스턴스 타입 비교 벤치마크
이전에 Vultr 벤치마크를 돌려보고 타 클라우드 서비스 성능이 궁금하기도 했고, 최근에 잠깐 iwinv 인스턴스를 생성해서 잠깐 사용하는 김에 벤치를 돌려봤는데 의외로(?) 점수가 꽤 잘 나와서 아
blog.sysbot32.com
대상
- High Performance - Gen2 Dedicated
- 2 vCPUs, 4GB, 41,800원/월 (hpa_2.4-g2)
이전과 마찬가지로 KR1-Z01, Ubuntu 24.04 LTS에서 진행했다.
CPU 모델 확인
processor : 0
vendor_id : AuthenticAMD
cpu family : 26
model : 68
model name : AMD EPYC 4545P 16-Core Processor
stepping : 0
microcode : 0xb40401c
cpu MHz : 2994.374
cache size : 512 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 16
wp : yes
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 pdpe1gb rdtscp lm rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust 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 vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b overflow_recov succor fsrm avx512_vp2intersect flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass ibpb_no_ret
bogomips : 5988.74
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
processor : 1
vendor_id : AuthenticAMD
cpu family : 26
model : 68
model name : AMD EPYC 4545P 16-Core Processor
stepping : 0
microcode : 0xb40401c
cpu MHz : 2994.374
cache size : 512 KB
physical id : 1
siblings : 1
core id : 0
cpu cores : 1
apicid : 1
initial apicid : 1
fpu : yes
fpu_exception : yes
cpuid level : 16
wp : yes
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 pdpe1gb rdtscp lm rep_good nopl cpuid extd_apicid tsc_known_freq pni pclmulqdq ssse3 fma cx16 sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw perfctr_core ssbd ibrs ibpb stibp ibrs_enhanced vmmcall fsgsbase tsc_adjust 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 vnmi avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg avx512_vpopcntdq rdpid movdiri movdir64b overflow_recov succor fsrm avx512_vp2intersect flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass ibpb_no_ret
bogomips : 5988.74
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
싱글 스레드 벤치마크
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 2540.76
General statistics:
total time: 3.9351s
total number of events: 10000
Latency (ms):
min: 0.39
avg: 0.39
max: 0.45
95th percentile: 0.39
sum: 3934.30
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 3.9343/0.00
멀티 스레드 벤치마크 (2)
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 2
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 4974.11
General statistics:
total time: 2.0098s
total number of events: 10000
Latency (ms):
min: 0.39
avg: 0.40
max: 0.80
95th percentile: 0.42
sum: 4017.30
Threads fairness:
events (avg/stddev): 5000.0000/109.00
execution time (avg/stddev): 2.0086/0.00
멀티 스레드 벤치마크 (4)
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 4
Initializing random number generator from current time
Prime numbers limit: 20000
Initializing worker threads...
Threads started!
CPU speed:
events per second: 4984.42
General statistics:
total time: 2.0056s
total number of events: 10000
Latency (ms):
min: 0.39
avg: 0.80
max: 6.22
95th percentile: 2.43
sum: 8004.39
Threads fairness:
events (avg/stddev): 2500.0000/41.02
execution time (avg/stddev): 2.0011/0.00
메모리 읽기 테스트
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
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: 104857600 (14003909.04 per second)
102400.00 MiB transferred (13675.69 MiB/sec)
General statistics:
total time: 7.4866s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.04
95th percentile: 0.00
sum: 2901.57
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 2.9016/0.00
메모리 쓰기 테스트
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
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: 104857600 (11358955.11 per second)
102400.00 MiB transferred (11092.73 MiB/sec)
General statistics:
total time: 9.2306s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.08
95th percentile: 0.00
sum: 4435.83
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 4.4358/0.00
디스크 테스트
sysbench 1.0.20 (using system LuaJIT 2.1.0-beta3)
Running the test with following options:
Number of threads: 1
Initializing random number generator from current time
Extra file open flags: (none)
128 files, 64MiB each
8GiB total file size
Block size 16KiB
Number of IO requests: 0
Read/Write ratio for combined random IO test: 1.50
Periodic FSYNC enabled, calling fsync() each 100 requests.
Calling fsync() at the end of test, Enabled.
Using synchronous I/O mode
Doing random r/w test
Initializing worker threads...
Threads started!
File operations:
reads/s: 871.52
writes/s: 581.01
fsyncs/s: 1859.37
Throughput:
read, MiB/s: 13.62
written, MiB/s: 9.08
General statistics:
total time: 300.0754s
total number of events: 993692
Latency (ms):
min: 0.00
avg: 0.30
max: 88.35
95th percentile: 1.70
sum: 299789.43
Threads fairness:
events (avg/stddev): 993692.0000/0.00
execution time (avg/stddev): 299.7894/0.00
결론
| hpa_2.4-g2 | |
| 명시된 스펙 | 2 vCPUs, 4GB |
| CPU 모델 | AMD EPYC 4545P 16-Core Processor |
| 싱글 스레드 | 2540.76 |
| 멀티 스레드 (2) | 4974.11 |
| 멀티 스레드 (4) | 4984.42 |
| 메모리 읽기 | 13675.69 |
| 메모리 쓰기 | 11092.73 |
| 디스크 | 13.62 9.08 |
| 가격 | 41,800원/월 |
이전 세대 High Performance에 비하면 성능이 굉장히 좋아졌다!
하지만, General Purpose와의 가격 차이에 비하면 성능 차이는 크지 않은 것으로 보인다.
서버용 CPU, 10Gbps가 필요하다 👉 High Performance Gen 2
일반 CPU, 2.5Gbps로도 충분하다 👉 General Purpose
인터넷 속도 테스트
인터넷 속도 테스트는 외부 영향을 많이 받아서 테스트 결과가 항상 일정하게 나온다는 보장이 없어서 그동안 안 했는데, 이번에는 번외로 한 번 돌려보고자 한다.
ping 테스트
AWS 리전별 S3 엔드포인트에 ping을 10번씩 날렸다.
🇰🇷 서울
PING s3.ap-northeast-2.amazonaws.com (3.5.189.41) 56(84) bytes of data.
64 bytes from 3.5.189.41: icmp_seq=1 ttl=53 time=2.52 ms
64 bytes from 3.5.189.41: icmp_seq=2 ttl=53 time=2.40 ms
64 bytes from 3.5.189.41: icmp_seq=3 ttl=53 time=2.42 ms
64 bytes from 3.5.189.41: icmp_seq=4 ttl=53 time=2.50 ms
64 bytes from 3.5.189.41: icmp_seq=5 ttl=53 time=2.43 ms
64 bytes from 3.5.189.41: icmp_seq=6 ttl=53 time=2.42 ms
64 bytes from 3.5.189.41: icmp_seq=7 ttl=53 time=2.45 ms
64 bytes from 3.5.189.41: icmp_seq=8 ttl=53 time=2.45 ms
64 bytes from 3.5.189.41: icmp_seq=9 ttl=53 time=2.44 ms
64 bytes from 3.5.189.41: icmp_seq=10 ttl=53 time=2.43 ms
--- s3.ap-northeast-2.amazonaws.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9014ms
rtt min/avg/max/mdev = 2.402/2.446/2.521/0.035 ms
🇯🇵 도쿄
PING s3.ap-northeast-1.amazonaws.com (3.5.155.26) 56(84) bytes of data.
64 bytes from 3.5.155.26: icmp_seq=1 ttl=53 time=35.3 ms
64 bytes from 3.5.155.26: icmp_seq=2 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=3 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=4 ttl=53 time=35.4 ms
64 bytes from 3.5.155.26: icmp_seq=5 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=6 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=7 ttl=53 time=35.3 ms
64 bytes from 3.5.155.26: icmp_seq=8 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=9 ttl=53 time=35.1 ms
64 bytes from 3.5.155.26: icmp_seq=10 ttl=53 time=35.2 ms
--- s3.ap-northeast-1.amazonaws.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9012ms
rtt min/avg/max/mdev = 35.107/35.201/35.421/0.102 ms
🇺🇸 버지니아 북부
PING s3.us-east-1.amazonaws.com (16.15.253.255) 56(84) bytes of data.
64 bytes from 16.15.253.255: icmp_seq=1 ttl=44 time=191 ms
64 bytes from 16.15.253.255: icmp_seq=2 ttl=44 time=191 ms
64 bytes from 16.15.253.255: icmp_seq=3 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=4 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=5 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=6 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=7 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=8 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=9 ttl=44 time=194 ms
64 bytes from 16.15.253.255: icmp_seq=10 ttl=44 time=194 ms
--- s3.us-east-1.amazonaws.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9012ms
rtt min/avg/max/mdev = 190.817/193.592/194.445/1.366 ms
🇺🇸 캘리포니아 북부
PING s3.us-west-1.amazonaws.com (16.15.0.167) 56(84) bytes of data.
64 bytes from 16.15.0.167: icmp_seq=1 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=2 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=3 ttl=51 time=131 ms
64 bytes from 16.15.0.167: icmp_seq=4 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=5 ttl=51 time=131 ms
64 bytes from 16.15.0.167: icmp_seq=6 ttl=51 time=131 ms
64 bytes from 16.15.0.167: icmp_seq=7 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=8 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=9 ttl=51 time=134 ms
64 bytes from 16.15.0.167: icmp_seq=10 ttl=51 time=131 ms
--- s3.us-west-1.amazonaws.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9013ms
rtt min/avg/max/mdev = 130.612/132.932/134.498/1.842 ms
🇸🇬 싱가포르
PING s3.ap-southeast-1.amazonaws.com (3.5.151.68) 56(84) bytes of data.
64 bytes from 3.5.151.68: icmp_seq=1 ttl=48 time=66.9 ms
64 bytes from 3.5.151.68: icmp_seq=2 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=3 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=4 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=5 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=6 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=7 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=8 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=9 ttl=48 time=66.8 ms
64 bytes from 3.5.151.68: icmp_seq=10 ttl=48 time=66.8 ms
--- s3.ap-southeast-1.amazonaws.com ping statistics ---
10 packets transmitted, 10 received, 0% packet loss, time 9007ms
rtt min/avg/max/mdev = 66.766/66.798/66.933/0.046 ms
다운로드 테스트
원래 speedtest-cli 쓰려고 했으나, 계속 에러나서 그냥 Vultr 속도 테스트용 파일 다운 받아 보는 걸로.
🇰🇷 서울
--2026-04-18 11:56:06-- https://sel-kor-ping.vultr.com/vultr.com.1000MB.bin
Resolving sel-kor-ping.vultr.com (sel-kor-ping.vultr.com)... 141.164.34.61
Connecting to sel-kor-ping.vultr.com (sel-kor-ping.vultr.com)|141.164.34.61|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1048576000 (1000M) [application/octet-stream]
Saving to: ‘vultr.cohttp://m.1000MB.bin’
vultr.cohttp://m.1000MB.bi 100%[===================>] 1000M 515MB/s in 1.9s
2026-04-18 11:56:08 (515 MB/s) - ‘vultr.cohttp://m.1000MB.bin’ saved [1048576000/1048576000]
🇯🇵 도쿄
--2026-04-18 11:57:24-- https://hnd-jp-ping.vultr.com/vultr.com.1000MB.bin
Resolving hnd-jp-ping.vultr.com (hnd-jp-ping.vultr.com)... 108.61.201.151
Connecting to hnd-jp-ping.vultr.com (hnd-jp-ping.vultr.com)|108.61.201.151|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1048576000 (1000M) [application/octet-stream]
Saving to: ‘vultr.cohttp://m.1000MB.bin.1’
vultr.cohttp://m.1000MB.bi 100%[===================>] 1000M 122MB/s in 9.2s
2026-04-18 11:57:33 (109 MB/s) - ‘vultr.cohttp://m.1000MB.bin.1’ saved [1048576000/1048576000]
🇺🇸 뉴욕
--2026-04-18 12:01:03-- https://nj-us-ping.vultr.com/vultr.com.1000MB.bin
Resolving nj-us-ping.vultr.com (nj-us-ping.vultr.com)... 108.61.149.182
Connecting to nj-us-ping.vultr.com (nj-us-ping.vultr.com)|108.61.149.182|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1048576000 (1000M) [application/octet-stream]
Saving to: ‘vultr.cohttp://m.1000MB.bin.2’
vultr.cohttp://m.1000MB.bi 100%[===================>] 1000M 5.98MB/s in 83s
2026-04-18 12:02:27 (12.0 MB/s) - ‘vultr.cohttp://m.1000MB.bin.2’ saved [1048576000/1048576000]
🇺🇸 실리콘밸리 (산호세)
--2026-04-18 12:02:51-- https://sjo-ca-us-ping.vultr.com/vultr.com.1000MB.bin
Resolving sjo-ca-us-ping.vultr.com (sjo-ca-us-ping.vultr.com)... 104.156.230.107
Connecting to sjo-ca-us-ping.vultr.com (sjo-ca-us-ping.vultr.com)|104.156.230.107|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1048576000 (1000M) [application/octet-stream]
Saving to: ‘vultr.cohttp://m.1000MB.bin.3’
vultr.cohttp://m.1000MB.bi 100%[===================>] 1000M 21.6MB/s in 48s
2026-04-18 12:03:39 (20.9 MB/s) - ‘vultr.cohttp://m.1000MB.bin.3’ saved [1048576000/1048576000]
🇸🇬 싱가포르
--2026-04-18 12:05:16-- https://sgp-ping.vultr.com/vultr.com.1000MB.bin
Resolving sgp-ping.vultr.com (sgp-ping.vultr.com)... 45.32.100.168
Connecting to sgp-ping.vultr.com (sgp-ping.vultr.com)|45.32.100.168|:443... connected.
HTTP request sent, awaiting response... 200 OK
Length: 1048576000 (1000M) [application/octet-stream]
Saving to: ‘vultr.cohttp://m.1000MB.bin.4’
vultr.cohttp://m.1000MB.bi 100%[===================>] 1000M 33.4MB/s in 97s
2026-04-18 12:06:53 (10.3 MB/s) - ‘vultr.cohttp://m.1000MB.bin.4’ saved [1048576000/1048576000]
'클라우드' 카테고리의 다른 글
| Vultr HDD 블록 스토리지 벤치마크 (3) | 2025.08.11 |
|---|---|
| iwinv 인스턴스 타입 비교 벤치마크 (2) | 2025.05.24 |
| Vultr 인스턴스 타입 비교 벤치마크 (0) | 2025.03.13 |