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 아예 정식으로 비교를 해보려고 한다.
2025.03.13 - [클라우드] - Vultr 인스턴스 타입 비교 벤치마크
Vultr 인스턴스 타입 비교 벤치마크
Vultr에는 인스턴스 타입이 여러 종류 있는데, 어떤 것을 선택해야 유리한지 판단하기 위해 알아보자. 비교 대상비교하려는 대상은 다음과 같다.Shared - Regular Performance2 vCPUs, 4 GB, $20.00 /mo (vc2-2c-4gb)
blog.sysbot32.com
비교 대상
- General Purpose - Shared
- 4 vCPU, 4 GB, 24,000원 /월 (vgna_4_n)
- General Purpose - Dedicated
- 2 vCPU, 4GB, 27,800원 /월 (gna_2.4_n)
- High Performance - Dedicated
- 2 vCPU, 4 GB, 34,900원 /월 (hpa_2.4_n)
SSD 서버는 이전 세대의 인스턴스 타입으로 현 시점에서는 선택할 이유가 없는 것으로 보여 제외했다. NVMe 서버 대비 가격이 딱히 저렴하지도 않다.
ZONE은 KR1-Z01, 운영체제는 Ubuntu 24.04 LTS로 모두 동일하게 진행했다.
CPU 모델 확인
General Purpose는 최대 5.1GHz의 고성능 AMD 프로세서, High Performance는 최대 3.3GHz의 서버용 AMD 프로세서를 사용한다고 하는데 확인해보자.
cat /proc/cpuinfo
vgna_4_n
processor : 0
vendor_id : AuthenticAMD
cpu family : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.836
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
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 : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.836
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
processor : 2
vendor_id : AuthenticAMD
cpu family : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.836
cache size : 512 KB
physical id : 2
siblings : 1
core id : 0
cpu cores : 1
apicid : 2
initial apicid : 2
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
processor : 3
vendor_id : AuthenticAMD
cpu family : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.836
cache size : 512 KB
physical id : 3
siblings : 1
core id : 0
cpu cores : 1
apicid : 3
initial apicid : 3
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
gna_2.4_n
processor : 0
vendor_id : AuthenticAMD
cpu family : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.838
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
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 : 25
model : 116
model name : AMD Ryzen 7 7840HS w/ Radeon 780M Graphics
stepping : 1
microcode : 0xa704107
cpu MHz : 3792.838
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 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 overflow_recov succor fsrm flush_l1d arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass srso
bogomips : 7585.67
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
General Purpose는 일반 PC용 CPU를 사용하고, Shared/Dedicated 여부만 차이가 있고 동일한 CPU를 사용하는 것으로 보인다.
hpa_2.4_n
processor : 0
vendor_id : AuthenticAMD
cpu family : 23
model : 49
model name : AMD EPYC 7352 24-Core Processor
stepping : 0
microcode : 0x8301055
cpu MHz : 2295.684
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 vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr wbnoinvd arat npt lbrv nrip_save tsc_scale vmcb_clean pausefilter pfthreshold v_vmsave_vmload vgif umip rdpid arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass retbleed smt_rsb srso
bogomips : 4591.36
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 : 23
model : 49
model name : AMD EPYC 7352 24-Core Processor
stepping : 0
microcode : 0x8301055
cpu MHz : 2295.684
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 vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr wbnoinvd arat npt lbrv nrip_save tsc_scale vmcb_clean pausefilter pfthreshold v_vmsave_vmload vgif umip rdpid arch_capabilities
bugs : sysret_ss_attrs spectre_v1 spectre_v2 spec_store_bypass retbleed smt_rsb srso
bogomips : 4591.36
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 48 bits physical, 48 bits virtual
power management:
High Performance는 서버용 CPU를 사용하고 있다.
싱글 스레드 벤치마크
sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 run
vgna_4_n
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: 1066.50
General statistics:
total time: 9.3760s
total number of events: 10000
Latency (ms):
min: 0.43
avg: 0.94
max: 12.53
95th percentile: 0.70
sum: 9369.30
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 9.3693/0.00
gna_2.4_n
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: 2309.36
General statistics:
total time: 4.3297s
total number of events: 10000
Latency (ms):
min: 0.43
avg: 0.43
max: 1.59
95th percentile: 0.44
sum: 4328.58
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 4.3286/0.00
hpa_2.4_n
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: 578.84
General statistics:
total time: 17.2744s
total number of events: 10000
Latency (ms):
min: 1.69
avg: 1.73
max: 2.23
95th percentile: 1.76
sum: 17272.48
Threads fairness:
events (avg/stddev): 10000.0000/0.00
execution time (avg/stddev): 17.2725/0.00
High Performance라더니 오히려 점수가 낮게 나왔다.
멀티 스레드 벤치마크 (2)
sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 --threads=2 run
vgna_4_n
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: 2116.60
General statistics:
total time: 4.7242s
total number of events: 10000
Latency (ms):
min: 0.44
avg: 0.94
max: 11.75
95th percentile: 0.73
sum: 9437.31
Threads fairness:
events (avg/stddev): 5000.0000/5.00
execution time (avg/stddev): 4.7187/0.00
gna_2.4_n
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: 4609.72
General statistics:
total time: 2.1689s
total number of events: 10000
Latency (ms):
min: 0.43
avg: 0.43
max: 1.50
95th percentile: 0.44
sum: 4336.07
Threads fairness:
events (avg/stddev): 5000.0000/9.00
execution time (avg/stddev): 2.1680/0.00
hpa_2.4_n
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: 1155.31
General statistics:
total time: 8.6545s
total number of events: 10000
Latency (ms):
min: 1.70
avg: 1.73
max: 3.38
95th percentile: 1.76
sum: 17304.32
Threads fairness:
events (avg/stddev): 5000.0000/1.00
execution time (avg/stddev): 8.6522/0.00
멀티 스레드 벤치마크 (4)
vCPU가 4개짜리인 플랜이 있어서 4개로도 돌려봤다.
sysbench cpu --events=10000 --cpu-max-prime=20000 --time=0 --threads=4 run
vgna_4_n
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: 4096.90
General statistics:
total time: 2.4405s
total number of events: 10000
Latency (ms):
min: 0.44
avg: 0.97
max: 11.67
95th percentile: 0.72
sum: 9745.32
Threads fairness:
events (avg/stddev): 2500.0000/3.24
execution time (avg/stddev): 2.4363/0.00
Shared 4 vCPU <= Dedicated 2 vCPU
gna_2.4_n
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: 4605.22
General statistics:
total time: 2.1710s
total number of events: 10000
Latency (ms):
min: 0.42
avg: 0.87
max: 6.49
95th percentile: 2.43
sum: 8679.96
Threads fairness:
events (avg/stddev): 2500.0000/39.27
execution time (avg/stddev): 2.1700/0.00
hpa_2.4_n
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: 1157.17
General statistics:
total time: 8.6406s
total number of events: 10000
Latency (ms):
min: 1.70
avg: 3.45
max: 5.99
95th percentile: 3.75
sum: 34546.47
Threads fairness:
events (avg/stddev): 2500.0000/1.22
execution time (avg/stddev): 8.6366/0.00
메모리 읽기 테스트
sysbench --test=memory --memory-block-size=1K --memory-scope=global --memory-total-size=100G --memory-oper=read --time=0 run
vgna_4_n
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 (6481121.32 per second)
102400.00 MiB transferred (6329.22 MiB/sec)
General statistics:
total time: 16.1785s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 12.12
95th percentile: 0.00
sum: 5665.92
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 5.6659/0.00
gna_2.4_n
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 (14130154.60 per second)
102400.00 MiB transferred (13798.98 MiB/sec)
General statistics:
total time: 7.4202s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.17
95th percentile: 0.00
sum: 2546.53
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 2.5465/0.00
hpa_2.4_n
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 (7465299.73 per second)
102400.00 MiB transferred (7290.33 MiB/sec)
General statistics:
total time: 14.0444s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.28
95th percentile: 0.00
sum: 5469.45
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 5.4695/0.00
메모리 쓰기 테스트
sysbench --test=memory --memory-block-size=1K --memory-scope=global --memory-total-size=100G --memory-oper=write --time=0 run
vgna_4_n
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 (5282834.51 per second)
102400.00 MiB transferred (5159.02 MiB/sec)
General statistics:
total time: 19.8484s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 12.08
95th percentile: 0.00
sum: 8910.71
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 8.9107/0.00
gna_2.4_n
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 (11642367.34 per second)
102400.00 MiB transferred (11369.50 MiB/sec)
General statistics:
total time: 9.0059s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.07
95th percentile: 0.00
sum: 4017.19
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 4.0172/0.00
hpa_2.4_n
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 (5985196.60 per second)
102400.00 MiB transferred (5844.92 MiB/sec)
General statistics:
total time: 17.5181s
total number of events: 104857600
Latency (ms):
min: 0.00
avg: 0.00
max: 0.14
95th percentile: 0.00
sum: 8956.82
Threads fairness:
events (avg/stddev): 104857600.0000/0.00
execution time (avg/stddev): 8.9568/0.00
디스크 테스트
sysbench --test=fileio --file-total-size=8G prepare
sysbench fileio --file-total-size=8G --file-test-mode=rndrw --time=300 --max-requests=0 run
sysbench --test=fileio --file-total-size=8G cleanup
vgna_4_n
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: 596.78
writes/s: 397.86
fsyncs/s: 1273.23
Throughput:
read, MiB/s: 9.32
written, MiB/s: 6.22
General statistics:
total time: 300.1079s
total number of events: 680479
Latency (ms):
min: 0.00
avg: 0.44
max: 23.81
95th percentile: 3.19
sum: 299689.38
Threads fairness:
events (avg/stddev): 680479.0000/0.00
execution time (avg/stddev): 299.6894/0.00
gna_2.4_n
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: 684.54
writes/s: 456.36
fsyncs/s: 1460.60
Throughput:
read, MiB/s: 10.70
written, MiB/s: 7.13
General statistics:
total time: 300.0278s
total number of events: 780392
Latency (ms):
min: 0.00
avg: 0.38
max: 12.34
95th percentile: 2.03
sum: 299794.27
Threads fairness:
events (avg/stddev): 780392.0000/0.00
execution time (avg/stddev): 299.7943/0.00
hpa_2.4_n
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: 1661.19
writes/s: 1107.46
fsyncs/s: 3544.07
Throughput:
read, MiB/s: 25.96
written, MiB/s: 17.30
General statistics:
total time: 300.0198s
total number of events: 1893822
Latency (ms):
min: 0.00
avg: 0.16
max: 3.94
95th percentile: 0.58
sum: 299397.32
Threads fairness:
events (avg/stddev): 1893822.0000/0.00
execution time (avg/stddev): 299.3973/0.00
결론
| vgna_4_n | gna_2.4_n | hpa_2.4_n | |
| 명시된 스펙 | 4 vCPUs, 4 GB | 2 vCPUs, 4 GB | 2 vCPUs, 4 GB |
| CPU 모델 | AMD Ryzen 7 7840HS w/ Radeon 780M Graphics | AMD Ryzen 7 7840HS w/ Radeon 780M Graphics | AMD EPYC 7352 24-Core Processor |
| 싱글 스레드 | 1066.50 | 2309.36 | 578.84 |
| 멀티 스레드 (2) | 2116.60 | 4609.72 | 1155.31 |
| 멀티 스레드 (4) | 4096.90 | 4605.22 | 1157.17 |
| 메모리 읽기 | 6329.22 | 13798.98 | 7290.33 |
| 메모리 쓰기 | 5159.02 | 11369.50 | 5844.92 |
| 디스크 | 9.32 6.22 |
10.70 7.13 |
25.96 17.30 |
| 가격 | 24,000원 /월 | 27,800원 /월 | 34,900원 /월 |
다음은 각각 성능에 따른 순위 및 가성비를 고려한 추천 플랜 및 비추천 플랜이다.
싱글 스레드 성능
gna_2.4_n > vgna_4_n > hpa_2.4_n
멀티 스레드 성능
gna_2.4_n > vgna_4_n > hpa_2.4_n
CPU 성능은 싱글/멀티 스레드 상관 없이 High Performance 플랜보다 General Purpose 플랜이 더 잘 나온다. High Performance 플랜이 이름 값을 못하고 있다.
서버용 CPU가 아닌 일반 PC용 CPU를 사용하는 것도 상관 없다면 iwinv의 General Purpose는 정말 가성비 깡패다.
메모리 읽기 성능
gna_2.4_n > hpa_2.4_n > vgna_4_n
메모리 쓰기 성능
gna_2.4_n > hpa_2.4_n > vgna_4_n
디스크 성능
hpa_2.4_n > gna_2.4_n > vgna_4_n
NVMe를 사용한다더니 General Purpose 플랜은 그냥 일반 SATA SSD 수준 성능만 나온다. 😑
※ 본 글에서 사용한 명령어는 서버 성능측정 (벤치마크 방법)을 참고했다.
'클라우드' 카테고리의 다른 글
| Vultr 인스턴스 타입 비교 벤치마크 (0) | 2025.03.13 |
|---|