まずWSL1の状態でベンチを動かしたあと、WSL2に変換して再実行しました。
環境は古いノートPC＋SSDです。
WSL1
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: DESKTOP-MTEHMSF: GNU/Linux
OS: GNU/Linux -- 4.4.0-18917-Microsoft -- #1000-Microsoft Fri Jun 07 19:42:00 PST 2019
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (5002.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (5002.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (5002.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (5002.0 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
22:55:07 up 59 min, 0 users, load average: 0.52, 0.58, 0.59; runlevel
------------------------------------------------------------------------
Benchmark Run: Thu Jun 13 2019 22:55:07 - 23:23:21
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 34864987.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4361.4 MWIPS (9.9 s, 7 samples)
Execl Throughput 262.1 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 70776.2 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 18227.2 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 221085.4 KBps (30.0 s, 2 samples)
Pipe Throughput 217786.2 lps (10.0 s, 7 samples)
Pipe-based Context Switching 31918.0 lps (10.0 s, 7 samples)
Process Creation 429.6 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 688.9 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 126.3 lpm (60.6 s, 2 samples)
System Call Overhead 271501.6 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 34864987.6 2987.6
Double-Precision Whetstone 55.0 4361.4 793.0
Execl Throughput 43.0 262.1 61.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 70776.2 178.7
File Copy 256 bufsize 500 maxblocks 1655.0 18227.2 110.1
File Copy 4096 bufsize 8000 maxblocks 5800.0 221085.4 381.2
Pipe Throughput 12440.0 217786.2 175.1
Pipe-based Context Switching 4000.0 31918.0 79.8
Process Creation 126.0 429.6 34.1
Shell Scripts (1 concurrent) 42.4 688.9 162.5
Shell Scripts (8 concurrent) 6.0 126.3 210.5
System Call Overhead 15000.0 271501.6 181.0
========
System Benchmarks Index Score 195.9
------------------------------------------------------------------------
Benchmark Run: Thu Jun 13 2019 23:23:21 - 23:51:54
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 55181887.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 11634.6 MWIPS (9.8 s, 7 samples)
Execl Throughput 472.1 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 51705.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 13391.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 171441.3 KBps (30.0 s, 2 samples)
Pipe Throughput 413731.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 123267.0 lps (10.0 s, 7 samples)
Process Creation 675.0 lps (30.1 s, 2 samples)
Shell Scripts (1 concurrent) 813.5 lpm (60.1 s, 2 samples)
Shell Scripts (8 concurrent) 103.9 lpm (61.2 s, 2 samples)
System Call Overhead 597742.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 55181887.6 4728.5
Double-Precision Whetstone 55.0 11634.6 2115.4
Execl Throughput 43.0 472.1 109.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 51705.7 130.6
File Copy 256 bufsize 500 maxblocks 1655.0 13391.1 80.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 171441.3 295.6
Pipe Throughput 12440.0 413731.6 332.6
Pipe-based Context Switching 4000.0 123267.0 308.2
Process Creation 126.0 675.0 53.6
Shell Scripts (1 concurrent) 42.4 813.5 191.9
Shell Scripts (8 concurrent) 6.0 103.9 173.2
System Call Overhead 15000.0 597742.5 398.5
========
System Benchmarks Index Score 281.7
WSL2
========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)
System: DESKTOP-MTEHMSF: GNU/Linux
OS: GNU/Linux -- 4.19.43-microsoft-standard -- #1 SMP Mon May 20 19:35:22 UTC 2019
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap="UTF-8", collate="UTF-8")
CPU 0: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (4988.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 1: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (4988.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 2: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (4988.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
CPU 3: Intel(R) Core(TM) i5-2450M CPU @ 2.50GHz (4988.7 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET
02:12:55 up 0 min, 0 users, load average: 0.04, 0.01, 0.00; runlevel
------------------------------------------------------------------------
Benchmark Run: Fri Jun 14 2019 02:12:55 - 02:41:01
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 33880708.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 4313.9 MWIPS (9.9 s, 7 samples)
Execl Throughput 3791.7 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 530306.9 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 104802.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1575922.7 KBps (30.0 s, 2 samples)
Pipe Throughput 790148.5 lps (10.0 s, 7 samples)
Pipe-based Context Switching 11518.4 lps (10.0 s, 7 samples)
Process Creation 4462.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 7035.5 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1574.0 lpm (60.0 s, 2 samples)
System Call Overhead 570331.7 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 33880708.3 2903.2
Double-Precision Whetstone 55.0 4313.9 784.4
Execl Throughput 43.0 3791.7 881.8
File Copy 1024 bufsize 2000 maxblocks 3960.0 530306.9 1339.2
File Copy 256 bufsize 500 maxblocks 1655.0 104802.5 633.2
File Copy 4096 bufsize 8000 maxblocks 5800.0 1575922.7 2717.1
Pipe Throughput 12440.0 790148.5 635.2
Pipe-based Context Switching 4000.0 11518.4 28.8
Process Creation 126.0 4462.8 354.2
Shell Scripts (1 concurrent) 42.4 7035.5 1659.3
Shell Scripts (8 concurrent) 6.0 1574.0 2623.4
System Call Overhead 15000.0 570331.7 380.2
========
System Benchmarks Index Score 778.6
------------------------------------------------------------------------
Benchmark Run: Fri Jun 14 2019 02:41:01 - 03:09:25
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 40715860.9 lps (10.0 s, 7 samples)
Double-Precision Whetstone 11363.7 MWIPS (10.1 s, 7 samples)
Execl Throughput 7159.2 lps (29.6 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 654775.8 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 175768.1 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1960394.1 KBps (30.0 s, 2 samples)
Pipe Throughput 1680014.6 lps (10.0 s, 7 samples)
Pipe-based Context Switching 241063.7 lps (10.0 s, 7 samples)
Process Creation 12042.2 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 10894.4 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 1537.7 lpm (60.2 s, 2 samples)
System Call Overhead 1492058.5 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 40715860.9 3488.9
Double-Precision Whetstone 55.0 11363.7 2066.1
Execl Throughput 43.0 7159.2 1664.9
File Copy 1024 bufsize 2000 maxblocks 3960.0 654775.8 1653.5
File Copy 256 bufsize 500 maxblocks 1655.0 175768.1 1062.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 1960394.1 3380.0
Pipe Throughput 12440.0 1680014.6 1350.5
Pipe-based Context Switching 4000.0 241063.7 602.7
Process Creation 126.0 12042.2 955.7
Shell Scripts (1 concurrent) 42.4 10894.4 2569.4
Shell Scripts (8 concurrent) 6.0 1537.7 2562.8
System Call Overhead 15000.0 1492058.5 994.7
========
System Benchmarks Index Score 1634.3
4多重のProcess Creationはたしかにおよそ20倍は早くなっていますね。
それと、4多重のFile Copy 256 bufsize 500 maxblocksが13倍ぐらいでしょうか？
一方で、1多重のPipe-based Context Switchingは1/3ぐらいに遅くなってたりもします。