Try to use raid and test bench mark

Are Enterprise Drives Worth The Cost?From a pure reliability perspective, the data we have says the answer is clear: No.If you’re OK with buying the replacements yourself after the warranty is up, then buy the cheaper consumer drives.Enterprise drives do have one advantage: longer warranties. That’s a benefit only if the higher price you pay for the longer warranty is less than what you expect to spend on replacing the drive.
R10 for any general purpose VMs and OS virtual disks. Generally, these are (given enough RAM) more skewed towards writes, and you will actually get the performance you need with fewer disks than R6
R6 for any file serving data. Generally skewed towards reads and capacity is more important than write performance generally.
R5 – only for SSDs and nonessential things. The risk is simply not worth the slight performance benefit over R6 and savings of one disk worth of space.

Base on the rule

I bought 4 DT01CAC300 to build a raid 5 to used as my main server staorage.

Server: 1U Supermicro Server X9SCL-F Intel Xeon E3-1220 V2 @ 3.10GHz Quad Core 32GB RAID 9750 RAID

Ivy Bridge Xeon E3-1220V2 22nm 4c4t 69w PCI 3.0 20lanes FCLGA1155,1600MHz DDR3 unbuffered ECC memory,Windows 7 Professional SP1,

Cinebench R11.5 Xeon E3-1220V2 5.61 Dual L5640 12.22 Dual X5670 16.96 i5-2500K 5.43 AMD-E 350 0.63
7ZIP AMD E-350 2315 E3-1220V2 13258 i5-2500K 13802 Dual L5640 37006 Dual X5670 50270
TrueCrypt AES AMD E-350 0.1GB/s E3-1220V2 2.5GB/s i5-2500K 2.6GB/2 Dual L5640 6.4GB/s Dual X5670 8.7GB/s
Handbrake 0.9.5 x264 Encoding(fps): AMD E-350 3 E3-1220V2 38 i5-2500K 40 Dual L5640 39 Dual X5670 42
Power comsumption(w) : AMD E-350 16-32 E3-1220V2 33-92 i5-2500K 32-122 Dual L5640 158-243 Dual X5670 174-378

E3-1230V2 is 4c8t and a little better

Dd using direct or synchronized I/O):
direct (use direct I/O for data)
dsync (use synchronized I/O for data)
sync (likewise, but also for metadata)

Server Throughput (Streaming I/O)
One gigabyte was written for the test:
dd if=/dev/zero of=/root/testfile bs=1G count=1 oflag=dsync
记录了1+0 的读入
记录了1+0 的写出
1073741824字节(1.1 GB)已复制,4.23159 秒,254 MB/秒

Server Latency
dd if=/dev/zero of=/root/testfile bs=512 count=1000 oflag=dsync
记录了1000+0 的读入
记录了1000+0 的写出
512000字节(512 kB)已复制,0.0812932 秒,6.3 MB/秒

Test result on a VM on another machine with one disk
dd if=/dev/zero of=~/testfile bs=1G count=1 oflag=direct
记录了1+0 的读入
记录了1+0 的写出
1073741824字节(1.1 GB)已复制,12.7396 秒,84.3 MB/秒

So Raid5 is 3 time fast than signle disk.

========================================================================
BYTE UNIX Benchmarks (Version 5.1.3)

System: centyp.szu.edu.cn: GNU/Linux
OS: GNU/Linux — 3.10.0-327.el7.x86_64 — #1 SMP Thu Nov 19 22:10:57 UTC 2015
Machine: x86_64 (x86_64)
Language: en_US.utf8 (charmap=”UTF-8″, collate=”UTF-8″)
CPU 0: Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz (6199.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 1: Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz (6199.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 2: Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz (6199.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
CPU 3: Intel(R) Xeon(R) CPU E3-1220 V2 @ 3.10GHz (6199.8 bogomips)
Hyper-Threading, x86-64, MMX, Physical Address Ext, SYSENTER/SYSEXIT, SYSCALL/SYSRET, Intel virtualization
15:22:49 up 5:53, 3 users, load average: 0.08, 0.05, 0.05; runlevel 2016-09-22

————————————————————————
Benchmark Run: 四 9月 22 2016 15:22:49 – 15:51:46
4 CPUs in system; running 1 parallel copy of tests

Dhrystone 2 using register variables 40889939.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 3309.3 MWIPS (15.1 s, 7 samples)
Execl Throughput 4263.5 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1287048.0 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 345209.0 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 3067480.5 KBps (30.0 s, 2 samples)
Pipe Throughput 1789495.0 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1.0 lps (10.0 s, 7 samples)
Process Creation 17709.8 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 9229.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3280.1 lpm (60.0 s, 2 samples)
System Call Overhead 2793342.6 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 40889939.3 3503.9
Double-Precision Whetstone 55.0 3309.3 601.7
Execl Throughput 43.0 4263.5 991.5
File Copy 1024 bufsize 2000 maxblocks 3960.0 1287048.0 3250.1
File Copy 256 bufsize 500 maxblocks 1655.0 345209.0 2085.9
File Copy 4096 bufsize 8000 maxblocks 5800.0 3067480.5 5288.8
Pipe Throughput 12440.0 1789495.0 1438.5
Pipe-based Context Switching 4000.0 1.0 0.0
Process Creation 126.0 17709.8 1405.5
Shell Scripts (1 concurrent) 42.4 9229.6 2176.8
Shell Scripts (8 concurrent) 6.0 3280.1 5466.9
System Call Overhead 15000.0 2793342.6 1862.2
========
System Benchmarks Index Score 671.4

————————————————————————
Benchmark Run: 四 9月 22 2016 15:51:46 – 16:20:10
4 CPUs in system; running 4 parallel copies of tests

Dhrystone 2 using register variables 154406222.7 lps (10.0 s, 7 samples)
Double-Precision Whetstone 16626.9 MWIPS (11.4 s, 7 samples)
Execl Throughput 20580.8 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 1868438.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 493678.3 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 5043699.0 KBps (30.0 s, 2 samples)
Pipe Throughput 6700204.4 lps (10.0 s, 7 samples)
Pipe-based Context Switching 1.0 lps (10.0 s, 7 samples)
Process Creation 65288.7 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 26584.9 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3657.4 lpm (60.0 s, 2 samples)
System Call Overhead 8619922.9 lps (10.0 s, 7 samples)

System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 154406222.7 13231.0
Double-Precision Whetstone 55.0 16626.9 3023.1
Execl Throughput 43.0 20580.8 4786.2
File Copy 1024 bufsize 2000 maxblocks 3960.0 1868438.5 4718.3
File Copy 256 bufsize 500 maxblocks 1655.0 493678.3 2983.0
File Copy 4096 bufsize 8000 maxblocks 5800.0 5043699.0 8696.0
Pipe Throughput 12440.0 6700204.4 5386.0
Pipe-based Context Switching 4000.0 1.0 0.0
Process Creation 126.0 65288.7 5181.6
Shell Scripts (1 concurrent) 42.4 26584.9 6270.0
Shell Scripts (8 concurrent) 6.0 3657.4 6095.7
System Call Overhead 15000.0 8619922.9 5746.6
========
System Benchmarks Index Score 1632.8

发表评论

电子邮件地址不会被公开。 必填项已用*标注