LinodeのDedicated CPUを借りてみました
· 約11分
2019-02-05からLinodeでDedicated CPUが選択可能になっていたので借りてベンチマークとか取ってみました
同じCPUが割り当てるということはなく、何種類かあるうちから割り当てられるようです。
JP2を選択、CPUはE5 2697 v4のようでした
/proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 79
model name : Intel(R) Xeon(R) CPU E5-2697 v4 @ 2.30GHz
stepping : 1
microcode : 0x1
cpu MHz : 2299.962
cache size : 16384 KB
physical id : 0
siblings : 1
core id : 0
cpu cores : 1
apicid : 0
initial apicid : 0
fpu : yes
fpu_exception : yes
cpuid level : 13
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 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology cpuid pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt arat
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf
bogomips : 4599.92
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
processor : 1
vendor_id : GenuineIntel
cpu family : 6
model : 79
model name : Intel(R) Xeon(R) CPU E5-2697 v4 @ 2.30GHz
stepping : 1
microcode : 0x1
cpu MHz : 2299.962
cache size : 16384 KB
physical id : 1
siblings : 1
core id : 0
cpu cores : 1
apicid : 1
initial apicid : 1
fpu : yes
fpu_exception : yes
cpuid level : 13
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 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl xtopology cpuid pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch invpcid_single pti fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm rdseed adx smap xsaveopt arat
bugs : cpu_meltdown spectre_v1 spectre_v2 spec_store_bypass l1tf
bogomips : 4599.92
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
UnixBench
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: localhost: GNU/Linux
OS: GNU/Linux -- 4.15.0-43-generic -- #46-Ubuntu SMP Thu Dec 6 14:45:28 UTC 2018
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Xeon(R) CPU E5-2697 v4 @ 2.30GHz (4599.9 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel(R) Xeon(R) CPU E5-2697 v4 @ 2.30GHz (4599.9 bogomips)
x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
06:40:27 up 3 min, 1 user, load average: 0.62, 0.57, 0.25; runlevel 2019-02-10
------------------------------------------------------------------------
Benchmark Run: Sun Feb 10 2019 06:40:27 - 07:09:05
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 32669027.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3891.4 MWIPS (13.2 s, 7 samples)
Execl Throughput 3137.8 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 579875.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 167926.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1449278.4 KBps (30.0 s, 2 samples)
Pipe Throughput 940623.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 183109.4 lps (10.0 s, 7 samples)
Process Creation 5712.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6593.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1177.3 lpm (60.0 s, 2 samples)
System Call Overhead 747077.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 32669027.9 2799.4
Double-Precision Whetstone 55.0 3891.4 707.5
Execl Throughput 43.0 3137.8 729.7
File Copy 1024 bufsize 2000 maxblocks 3960.0 579875.8 1464.3
File Copy 256 bufsize 500 maxblocks 1655.0 167926.0 1014.7
File Copy 4096 bufsize 8000 maxblocks 5800.0 1449278.4 2498.8
Pipe Throughput 12440.0 940623.1 756.1
Pipe-based Context Switching 4000.0 183109.4 457.8
Process Creation 126.0 5712.0 453.3
Shell Scripts (1 concurrent) 42.4 6593.0 1555.0
Shell Scripts (8 concurrent) 6.0 1177.3 1962.1
System Call Overhead 15000.0 747077.6 498.1
========
System Benchmarks Index Score 1020.7
------------------------------------------------------------------------
Benchmark Run: Sun Feb 10 2019 07:09:05 - 07:37:07
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 41602427.0 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8896.9 MWIPS (9.7 s, 7 samples)
Execl Throughput 4537.3 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 851625.6 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 238691.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2247450.9 KBps (30.0 s, 2 samples)
Pipe Throughput 1393582.9 lps (10.0 s, 7 samples)
Pipe-based Context Switching 278697.2 lps (10.0 s, 7 samples)
Process Creation 10646.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 8790.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1202.7 lpm (60.0 s, 2 samples)
System Call Overhead 1270597.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 41602427.0 3564.9
Double-Precision Whetstone 55.0 8896.9 1617.6
Execl Throughput 43.0 4537.3 1055.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 851625.6 2150.6
File Copy 256 bufsize 500 maxblocks 1655.0 238691.2 1442.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 2247450.9 3874.9
Pipe Throughput 12440.0 1393582.9 1120.2
Pipe-based Context Switching 4000.0 278697.2 696.7
Process Creation 126.0 10646.9 845.0
Shell Scripts (1 concurrent) 42.4 8790.2 2073.2
Shell Scripts (8 concurrent) 6.0 1202.7 2004.5
System Call Overhead 15000.0 1270597.6 847.1
========
System Benchmarks Index Score 1534.0
Dallasを選択、CPUはAMD EPYC 7501でした
/proc/cpuinfo
processor : 0
vendor_id : AuthenticAMD
cpu family : 23
model : 1
model name : AMD EPYC 7501 32-Core Processor
stepping : 2
microcode : 0x1000065
cpu MHz : 2000.002
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 : 13
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 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 cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ssbd ibpb vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt sha_ni xsaveopt xsavec xgetbv1 virt_ssbd arat
bugs : fxsave_leak sysret_ss_attrs null_seg spectre_v1 spectre_v2 spec_store_bypass
bogomips : 4000.00
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
processor : 1
vendor_id : AuthenticAMD
cpu family : 23
model : 1
model name : AMD EPYC 7501 32-Core Processor
stepping : 2
microcode : 0x1000065
cpu MHz : 2000.002
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 : 13
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 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 cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw ssbd ibpb vmmcall fsgsbase tsc_adjust bmi1 avx2 smep bmi2 rdseed adx smap clflushopt sha_ni xsaveopt xsavec xgetbv1 virt_ssbd arat
bugs : fxsave_leak sysret_ss_attrs null_seg spectre_v1 spectre_v2 spec_store_bypass
bogomips : 4000.00
TLB size : 1024 4K pages
clflush size : 64
cache_alignment : 64
address sizes : 40 bits physical, 48 bits virtual
power management:
UnixBench
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: localhost: GNU/Linux
OS: GNU/Linux -- 4.15.0-43-generic -- #46-Ubuntu SMP Thu Dec 6 14:45:28 UTC 2018
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: AMD EPYC 7501 32-Core Processor (4000.0 bogomips)
x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: AMD EPYC 7501 32-Core Processor (4000.0 bogomips)
x86-64, MMX, AMD MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
08:51:29 up 3 min, 1 user, load average: 0.42, 0.30, 0.12; runlevel 2019-02-10
------------------------------------------------------------------------
Benchmark Run: Sun Feb 10 2019 08:51:29 - 09:19:28
2 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 31094686.2 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4731.1 MWIPS (9.6 s, 7 samples)
Execl Throughput 3934.6 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 779144.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 220101.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1935516.4 KBps (30.0 s, 2 samples)
Pipe Throughput 1336429.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 163706.5 lps (10.0 s, 7 samples)
Process Creation 9365.5 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 9123.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1514.6 lpm (60.0 s, 2 samples)
System Call Overhead 2031584.4 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 31094686.2 2664.5
Double-Precision Whetstone 55.0 4731.1 860.2
Execl Throughput 43.0 3934.6 915.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 779144.0 1967.5
File Copy 256 bufsize 500 maxblocks 1655.0 220101.3 1329.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 1935516.4 3337.1
Pipe Throughput 12440.0 1336429.6 1074.3
Pipe-based Context Switching 4000.0 163706.5 409.3
Process Creation 126.0 9365.5 743.3
Shell Scripts (1 concurrent) 42.4 9123.9 2151.9
Shell Scripts (8 concurrent) 6.0 1514.6 2524.3
System Call Overhead 15000.0 2031584.4 1354.4
========
System Benchmarks Index Score 1370.8
------------------------------------------------------------------------
Benchmark Run: Sun Feb 10 2019 09:19:28 - 09:47:29
2 CPUs in system; running 2 parallel copies of tests
Dhrystone 2 using register variables 41932122.8 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8202.3 MWIPS (9.7 s, 7 samples)
Execl Throughput 5934.0 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1108519.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 312623.4 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 2820435.9 KBps (30.0 s, 2 samples)
Pipe Throughput 1931434.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 308361.0 lps (10.0 s, 7 samples)
Process Creation 18103.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 11035.0 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1537.3 lpm (60.0 s, 2 samples)
System Call Overhead 3152066.3 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 41932122.8 3593.2
Double-Precision Whetstone 55.0 8202.3 1491.3
Execl Throughput 43.0 5934.0 1380.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 1108519.5 2799.3
File Copy 256 bufsize 500 maxblocks 1655.0 312623.4 1889.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 2820435.9 4862.8
Pipe Throughput 12440.0 1931434.1 1552.6
Pipe-based Context Switching 4000.0 308361.0 770.9
Process Creation 126.0 18103.2 1436.8
Shell Scripts (1 concurrent) 42.4 11035.0 2602.6
Shell Scripts (8 concurrent) 6.0 1537.3 2562.2
System Call Overhead 15000.0 3152066.3 2101.4
========
System Benchmarks Index Score 2018.8
UnixBenchのスコアはXeon E5-2697 v4が1500なのに対してEPYC 7501が2000と結構な差があるようです。
また、今回は二回しかインスタンスガチャを回していないのでこれらのCPUしか引きませんでしたが、他にもあるようです(Xeon Gold 6148とか)
今回はJP2のリージョンの方が悪い結果でしたが、恐らく他のCPUもあると思われます。(同一リージョンでも複数種類あるようなので)