Home > Computer > Measure Actual Hard Disk Performance Under Windows

Measure Actual Hard Disk Performance Under Windows

Posted By Raymond In Category: Computer

Feb
28
2008
Donate

Did you know that your actual hard disk performance under Windows is determined not only by the speed of your harddrive (as measured by most disk benchmarks) but also by factors such as the mainboard chipset, UDMA/PIO mode, RAID configuration, CPU speed, RAM size and the quality of drivers? That’s true… It doesn’t mean that if you have a 7200RPM SATA hard drive, you will get the same hard disk performance as your friend who is also using a 7200RPM SATA hard drive but on a more powerful computer.

how to calculate hard drive performanceI knew about this logic all this while but never really put my desktop and laptop to the test. My desktop computer is quite a powerful one compared to my laptop but I was surprised to get better benchmark results for my laptop!


Before I show you the results of my hard disk drive benchmark and also the tools that I use, here is my desktop and laptop hardware configuration.

Laptop
- Intel Pentium M 1.73GHz (533MHz FSB, 2MB L2 cache)
- Intel 915PM Chipset
- 2GB DDR2-533MHz Memory (Dual Channel)
- 54000 RPM (IDE)

Desktop
- Intel Pentium D 2.8GHz (800MHz 1MB L2 cache)
- Intel 945P Chipset
- 1GB DDR2-667MHz Memory (Single Channel)
- 7200 RPM (SATA)

DiskTT simply tries to give you an indication of the amount of Megabytes per second read and write speed you get under your normal Windows 9x/ME/NT/2000/XP environment. DiskTT measures a disk’s throughput under 32-bit windows using Windows core read and write functions. DiskTT is small, free and portable. I’ve set the test file size as 1024MB.

Laptop hard drive performance with DiskTT.
laptop

Desktop hard drive performance with DiskTT.
desktop

As you can see, the 5400RPM hard drive on my laptop performs much better than my 7200RPM desktop hard drive. At first I didn’t believe because I always thought that my desktop is more powerful and faster than my laptop. So I copied an AVI movie file to both laptop and desktop. Then I manually copy the movie file to another location. My laptop was able to finish copying the file in half the time compared to my desktop. Is RAM really that important in hard disk performance? Please share with me your hard drive benchmark test results. Remember to set the size of test file to 1024MB.

[ Download Disk Throughput Tester ]


Related posts:
  • Boost Disk Drive Performance In Windows Vista
  • A FREE Hard Disk utility to Benchmark and Test your hard drive
  • Accelerate Hard Disk Drive to Speed Up File Transfer
  • Install XP: Setup did not find any hard disk drives installed in your computer
  • Find and Remove Duplicate Files to Free Up Hard Disk Space
    • Michael

      Work pc:
      AMD Sempron 3000+
      Nvidia Nforce 3 Chipset
      1GB DDR-333 Memory (Single Channel)
      7200 RPM (SATA)
      Write: 18.6MB per sec
      Read: 13.3MB per sec
      Random: 6.5MB per sec

    • Cute Lemon Jokes

      Huh ? really ? I always though that laptop cannot better than PC even the spec is higher. Today I learn some thing. So , not bad to buy a laptop ma.

    • Nufsed

      Hi,
      will disk throughput tester work with Vista?

    • exulter

      Here is my desktop hardware configuration..

      - Intel Pentium D (820) 2.8GHz – 800MHz, 1MB L2 cache
      - Intel 945P Chipset
      - Memory : 1GB DDR2-667MHz + 2x512MB DDR2-533MHz

      Here is the result using 1024MB test file:

      80GB HDD, 7200 RPM, 8MB Buffer, SATA:
      - Sequential write : 1024Mb in 33.81s = 30.3MB per second
      - Sequential read : 1024Mb in 1.59s = 642.8MB per second
      - random access : 1024Mb in 1.31s = 780.5MB per second

      250GB, 7200 RPM, 16MB Buffer, SATA2:
      - Sequential write : 1024Mb in 30.77s = 33.3MB per second
      - Sequential read : 1024Mb in 1.61s = 636.4MB per second
      - random access : 1024Mb in 1.28s = 799.4MB per second

    • pbrane

      Desktop:
      Vista 32bit
      C2D 4400@2,66
      Intel P35 Chipset
      2 GB DDR2-667@800

      WD 360GB, 7200 RPM, 8 MB Buffer, SATA2:
      1024Mb in 29.59s = 34.6MB per second
      1024Mb in 15.65s = 65.4MB per second
      1024Mb in 10.64s = 96.2MB per second

      Drive is 1/2 year old.

    • http://www.raymond.cc/ Raymond

      exulter, looks like your hard disk performs really well with your configuration.

      Anyone tested DiskTT on laptop?

    • Rishad

      exulter has crazy read and access, but writing is pretty slow.

    • rlecocq

      Has anyone considered wether or not a fragmented disk would effect the results?

      I would think so……

      Rob./

    • johnny

      why cant i post

    • shadowarrior

      My desktop:

      Intel C2D e6750 @ 2.66
      Intel X38
      Ram: 2Gb 1066 Dual Channel

      HD 1 : WD Raptor 36Gb

      1024Mb in 27.47s = 37.3MB per second
      1024Mb in 0.50s = 2048.0MB per second
      1024Mb in 0.64s = 1600.0MB per second

      HD 2 : WD Caviar 160Gb

      1024Mb in 25.38s = 40.4MB per second
      1024Mb in 1.53s = 668.4MB per second
      1024Mb in 1.69s = 606.6MB per second

    • http://free-iphone-apple.blogspot.com mark

      Yeah, it’s mainly b/c the memory is twice the size as the desktop (more data is copied to it first then written to your HD).

    • http://www.tlc.gr Teddie

      Hi Raymond. Always fun to read your blog.

      Here’s my results (dual boot desktop PC):

      CPU: Intel Pentium Q6600 Core 2 Quad
      2.4 Ghz (65nm – Kentsfield family)

      M/B: Gigabyte x38-DQ6 | Intel X38 Chipset

      RAM: 2 Gb DDR2 @ 800 Mhz
      (2 DIMMS x 1 Mb PC6400 – Dual Channel)

      HDD 1: Western Digital 320 Gb | SATA II
      16MB buffer | 7200 rpm (model WD3200KS)
      Running XP pro SP2 x86

      Sequential write: 1024Mb in 30.42s = 33.7MB per second
      Sequential read: 1024Mb in 1.69s = 607.0MB per second
      Random access: 1024Mb in 1.80s = 569.8MB per second

      HDD 2: Western Digital 500 Gb | SATA II
      16MB buffer | 7200 rpm (model WD5000AAKS)
      Running Vista Ultimate x64

      Sequential write: 1024Mb in 24.29s = 42.2MB per second
      Sequential read: 1024Mb in 1.64s = 623MB per second
      Random access: 1024Mb in 0.80s = 1286.4MB per second

    • John

      Windows xp pro sp2
      Pentium4 northwood 2.8ghz @ 3.4ghz
      Intel i865p chipset
      1.5gb DDR 200mhz

      WD Caviar SE16 250Gb sata 16Mb Buffer 7200rpm
      1024Mb in 29.58s = 34.6MB per second
      1024Mb in 2.25s = 455.1MB per second
      1024Mb in 2.05s = 500.2MB per second

    • Odie

      My Laptop (Acer 5315)
      Intel Celeron 530
      1.73 GHz, 533MHz FSB, 1MB L2 cache
      1 GB DDR2

      I have done 6 tests all with different results. The best was

      Sequential write 22,2 MB/s
      Sequential read 33,6 MB/s
      Random access 9,7 MB/s

      The worst was

      Sequential write 13,1 MB/s
      Sequential read 28,0 MB/s
      Random access 10,0 MB/s

      There obviously is something wrong on my machine but what?

    • FMD

      Laptop
      C2D 1.66GHz
      1.5GB RAM

      Sequential write 1024Mb in 70.12s = 14.6MB per second
      Sequential read 1024Mb in 41.75s = 24.5MB per second
      Random access 1024Mb in 4.11s = 249.2MB per second

    • http://free-iphone-apple.blogspot.com mark

      Odie: What kind of HD?

    • http://www.raymond.cc/ Raymond

      Odie, it might be some running programs on your computer causing different results each time you run a test.

    • Ann

      Intel Penitum 4 CPU 2.80 GHz
      2.79 GHz, 1G Ram
      Navida GeForce 7300 LE

      Write 1024 – 42.92s=23.9mb
      Read 1024 – 21.58s=47.5mb
      random 1024-66.75=15.3 mb

    • Odie

      Hi Mark

      I have moved this to the general forum

      My HDD is a Seagate Momentus 5400.3 (120GB)

    • Knysliux001

      Raymond, You are comparing SATA to IDE hard disks. They will never perform equal. Furthermore HDD in your laptop spins slower than in desktop PC. RPM really have some effect on speed :)
      Also I think that anti virus should be turned off while benchmarking (unless you are benchmarking the antivirus software speed).
      It’s interesting to compare different machines though.

    • abhishek

      Actually, There is one thing that even top physicists of todays modern science and which was eventally the dream of einstein is to understand and make out the UNIFIED FIELD THEORY, the same thing holds true in the case of IT, or computers, is to measure the performance of a system.

      There are too many factors which affects the performance of a system. This factors includes, the OD, the HDD, the processor, RAM and stuffs like that. But this factors also include the user and the way the user is using the computer.
      These factors interact with each other in an unique way.
      So, it is really a very difficult task to make out, how these factors interact with each other, so as to bring out the overall performance of the system.
      No matter how much tesing and benchmarking, we do, always a new result will be derived in a new day.
      This is my thought, what you guys think of this issueee???

    • exulter

      i tested about 4-5 times.. and my result still something like that.. why my write very slow?

    • bovine

      ok this program seems a bit on the “Lite” side – ditch it, use hdtach ffs

      i dont see how u choose which hard disk to test if u have more than one??!

      suggest u use hdtach instead (freeware)
      http://en.wikipedia.org/wiki/HD_Tach

      gives much more info

      fyi my results on asus a8v deluxe, 4600 dual core amd cpu samsung 500gb sata on the via raid controller in ide mode gives:

      sequential write 46.4mb sec
      read 612.0mb
      random 873.0 mb sec

      looks like the read and random must be cached?

      hdtach shows better info and u can choose drive:

      sequential read gives peak of 80 at start, and tails down to 40mb sec end of drive

      burst is 117mb

      random access 21 ms

      with all these apps u need to make sure u dont have any other windows apps running and use task manager to prune any TSRs

      i have utorrent going and stuff so results could be better

      enjoy, try hdtach seriously!

    • Pascal

      Windows vista Ultimate Sp1
      Intel pentium dual E2140 @ 1.60GHZ running at 2.88Ghz
      Gigabyte GA-X38-DQ6 (X38)
      Dual samsung 750GB 32MB SATA II SPINPOINT F1 (raid 0)

      sequential write 1024Mb in 17.38s = 58.9MB per second
      Sequential read 1024Mb in 0.84s = 1213.3MB per second
      Random access 1024Mb in 0.84s = 1213.3MB per second

    • http://prakharagrawal.wordpress.com/ Prakhar agrawal

      Hi Raymond,

      Here are my results:

      Intel c2d e6400 @ 2.57 Ghz
      Asus P5W DH Deluxe
      2 x 1 GB Ram 800 Mhz
      Seagate 250 GB Sata II

      1024Mb in 33.36s = 30.7MB per second
      1024Mb in 1.80s = 570.2MB per second
      1024Mb in 1.76s = 580.2MB per second

      Are these results low as compared to my config ?

    • http://marco maria

      hacker

    • bala

      hi Raymond,

      My desktop info:

      System:Microsoft windows xp Home edition
      Version 2002
      service pack 2

      Computer: AMD Sempron(tm) Processor
      3400+
      1.80GHz, 512 MB of RAM
      Physical Address Extension

      These are my results for the test:

      Sequential write:

      I did it on my desktop and got this:
      1204Mb in 38.03s = 31.7MB per second
      1204Mb in 28.19s = 42.7MB per second
      1204Mb in 139.84s = 8.6MB per second

      Raymond, is my com slow? I find it is extremely laggy… need your help on this. Sorry for all the questions i ask… But thanks for the tips. They always work.

    • http://idonthave bond

      respected sir,
      i am teaching some computer education related as windows xp,server, means i am persuing MCSE….
      i need ur help about the how to hack the any other administrator password….. xp, server or any other…… so please… give some tricks and tips…..

    • John Doe

      Laptop
      Windows XP Pro
      SP 2

      Intel pentium M 1.6 GHz
      512 RAM

      Test Results:
      1024Mb in 157.27s = 6.5 MB per second
      1024Mb in 121.82s = 8.4 MB per second
      1024Mb in 293.93s = 2.5 MB per second

    • Chris K

      Server IBM x354
      Windows 2003SP2

      Intel XEON 3GHz
      2GB

      2 * LACIE Biggest SATAII RAID 10 & Windows Software Raid 1 (RAID 110)

      1129Mb in 24.41s = 46.3MB per second
      1129Mb in 1.84s = 612.3MB per second
      1129Mb in 1.53s = 737.4MB per second

    • Flame

      Hi guys, please do not overestimate this old proggie and it’s results, nowadays’ drives have extremely optimised caching strategies that “fool” this program’s measurements. It just writes a _file_ to your drive, then works with it the same way like any other program on your computer. Select 1GB size or more to get possibly more accurate results. :)
      And yes, volume fragmentation matters pretty much, as is the positioning of test file on the drive . :)

    • http://www.ugosweb.com ugosweb

      I tried with my Laptodp HD and with my USB HD.
      I\’m really surpised to see that the external USB HD is a little better then the internal HD.

      my laptop is:
      DELL Latitude D820
      Core Duo T5600 1,83GHz
      2GB RAM

      Internal HD:
      120 GB
      5400 RPM

      write: 1024Mb in 51.84s = 19.8MB per second
      read: 1024Mb in 2.58s = 397.2MB per second
      random: 1024Mb in 3.00s = 341.3MB per second

      extednal USB HD:
      Fujitsu
      250 MB
      SATA
      USB2.0

      write: 1024Mb in 45.95s = 22.3MB per second
      read: 1024Mb in 2.61s = 392.5MB per second
      random: 1024Mb in 2.98s = 343.0MB per second

    • http://www.raymond.cc/ Raymond

      An update. I updated my desktop to 2GB (1GB + 1GB) dual channel.

      1024Mb in 39.55s = 25.9MB per second
      1024Mb in 1.58s = 648.9MB per second
      1024Mb in 1.31s = 780.5MB per second

      As you can see, the sequential read and random access had a huge improvement.

    • Jonney27

      Laptop
      AMD Turion 64*2 Mobile technology
      TL (1.6Ghz, 256kb + 256kb L2 cash, up to 1600MHz system bus)
      Windows XP
      1GB RAM
      120GB (5400rmp) hard drive
      NVIDIA GeForce Go 6150 graphics with up to 128MB shared video memory

      Test results:
      Write: 15.3 MB/s
      Read: 32.2 Mb/s
      Random access: 6.6 Mb/s

    • irjan

      Thanks Ray!!!

    • http://www.housethatcodebuilt.com Jankos

      Desktop:
      core 2 duo e6750 @ 2.66Ghz
      4GB DDR2 1066Mhz
      500GB WD5000AAKS

      write: 1024Mb in 19.36s = 52.9MB per second
      read: 1024Mb in 0.64s = 1600.0MB per second
      random: 1024Mb in 0.72s = 1428.2MB per second

    • http://www.raymond.cc/ Raymond

      Seems like the more RAM you got, the faster the read and random access.

    • jpdamigaman

      pentium d 3.0 GHZ
      1 gig ddr dual channel mode
      80 gig HD few mths old

      SEQ WRITE 1024Mb in 33.61s = 30.5MB per second
      SEQ READ 1024Mb in 20.52s = 49.9MB per second
      RANDOM ACC 1024Mb in 55.31s = 18.5MB per second

    • King Arthas

      Desktop:
      AMD Athlon 64 X2 3600+ Processor
      Sapphire PURE PE-AM2RS485M Mainboard
      Hynix 2 GB in dual channel mode
      160 GB SATA2 Hitachi Deskstar – 2 yrs old

      Sequential Write: 1024Mb in 33.84s = 30.3MB per second
      Sequential Read: 1024Mb in 1.50s = 682.7MB per second
      Random Access: 1024Mb in 1.91s = 537.0MB per second

    • tab

      Desktop:
      Intel Q6600 @ 3.6GHz
      Asus P5E
      Kingston 4GB
      1 * Seagate 500 GB (ST3500320AS)
      2 * 500 GB WD (WD5000AAKS) in Raid 0
      Vista x64

      Sequential Write: 1024Mb in 21.86s = 46.9MB per second
      Sequential Read: 1024Mb in 0.50s = 2048.0MB per second
      Random Access: 1024Mb in 0.48s = 2115.7MB per second

    • Arzie

      i have only 512MB of ram and my HDD is running terribly slow! about 2.2 MB/s

      any suggestions?

    • P_Dev

      Desktop:
      Intel Q6600 @ 3.7GHz
      Asus P5K Premium
      Patriot Viper Fin 8GB
      2 * 300 GB WD Velociraptor (WD3000GLFS) in Raid 0
      Vista x64

      Sequential Write: 1024Mb in 10.08s = 101.6MB per second
      Sequential Read: 1024Mb in 0.56s = 1825.3MB per second
      Random Access: 1024Mb in 0.89s = 1150.6MB per second

    • Rahul

      P4 631 – 3 GHz
      2*1GB Corsair @ 667
      Nvidia 7300 GS
      Sony 18X DVD-SATA

      HD1: Seagate. 160 GB. SATA 300.
      Sequential write: 128Mb = 3.2MB per second
      Sequential read: 128Mb = 627per second
      Random access: 128Mb = 820MB per second

      HD2: Seagate. 400 GB. SATA 300.
      Sequential write: 128Mb = 36MB per second
      Sequential read: 128Mb = 630MB per second
      Random access: 128Mb = 744MB per second

      lovetocommunicate@gmail.com

    • Acidvertigo79

      AMD ATHLON 5600+ X2
      2*1GB Team Value
      ABIT FATAL1TY AN9 32X
      NFORCE 590 SLI
      Nvidia 8800 GTS 320Mb
      250GB HITACHI TRAVELSTAR HDT725025VLA SATA II

      Sequential write: 1024Mb = 35.3MB per second
      Sequential read: 1024Mb = 57.4MB per second
      Random access: 1024Mb = 1170.3MB per second

    • http://www.s-anand.net/ S Anand

      Dell Latitude D420 laptop:

      Sequential write: 10.4MB per second
      Sequential write: 16.1MB per second
      Random access: 5.6MB per second

    • Knowland

      Intel Q9450
      8GB DDR2 800
      4 * WD 1600YS Raid Edition
      3ware 9650SE 4 channel @ raid 0
      No Paging File set

      Seq Write: 1024Mb in 10.42s = 98.3MB per second
      Seq Read: 1024Mb in 0.70s = 1458.7MB per second
      Random: 1024Mb in 0.94s = 1094.0MB per second

      I had a file downloading and was running WMP :\ :P

      Surprised that two WD VelociRaptors beat 4 SATA II 7200 drives. They must be DA SHIZNIT

    • jing

      I am using a Windows XP VMware ESX guest running on a iSCSI disk. I got a pathetic 3.8 mb / sec

    • GLUtoni

      Did some tweakin with Intel matrix storage caches as my write speed seemed way too slow. Now it\’s better but not as good I wished for maybe occing my RAM to 1066 will improve it.

      OS: Vista 64
      MBO P5Q-E, CPU E8500 @ 3.16Ghz, 8Gb OCZ DDR2@800Ghz, HDD 2 x WD6400AAKS on Intel P45 / ICH10R set as RAID0

      Only volume write back cache in use

      1024Mb in 214.88s = 4.8MB per second
      1024Mb in 0.86s = 1193.5MB per second
      1024Mb in 0.94s = 1094.0MB per second

      Hard drive data caches enabled 2 x 16Mb and volume write-back chache

      1024Mb in 23.21s = 44.1MB per second
      1024Mb in 0.81s = 1262.6MB per second
      1024Mb in 0.94s = 1094.0MB per second

      Hard drive data caches enabled 2 x 16Mb and volume write-back cache disabled

      1024Mb in 9.56s = 107.1MB per second
      1024Mb in 0.83s = 1238.2MB per second
      1024Mb in 0.95s = 1075.6MB per second

      funny that volume write back chache slows down writes that much or DTT and vita 64-bit just dont mix well.

    • Fred

      Questo e’ il thread piu’ vicino al problema che sto cercando di risolvere e di cui ancora non si trova nulla;
      in sostanza dopo una montagna di prove sembra che la maggior parte dei problemi sia da imputarsi a windows e da come gestisce le operazioni di multi i/o .
      Vorrei che faceste questa prova, soprattutto per chi dispone di sistemi sia a 32bit che a 64bit installati, ovvero provate sullo stesso a:
      avviare il defrag di windows (io uso jkdefrag.exe -a7 d: ma fa lo stesso), poi mentre il vostro defrag lavora copiate un file di grossa dimensione come una iso o dvdiso nella stessa partizione dove risiesiede il file o meglio su una partizione molto distante all’interno del disco, sono sicuro che ne vedrete di tutti i colori;
      dopo molti tweaks, cambio di controller sata, driver diversi, ahci o non ahci ho visto che mentre con i sistemi a 32bit si riesce ad ottenere risultati decenti, ovvero copia di file durante defrag da 8 a 25mb/s con i sistemi a 64bit la velocita’ ottenuta va a seconda delle configurazioni da 2 a 10mb/s e i dischi sembra ricordano i trapani col percussore a indicare che le prestazioni sul multi i/o sono disastrose e sembrea appunto essere un problema di come windows gestisce il filesystem.
      Vorrei sapere a vi cosa succede, i programmi comuni di test del disco infatti non sembrano comprendere una situazione di lavoro del genere e quindi non si riesco a fare dei rilievi validi.

    • Mr. Mentor

      Greetings,

      Repaired motherboard utilizing Celeron at 1482 MHz Remember, *ALWAYS* a capital \”H\” in frequency reference to pay homage to Mr. Heinrich Hertz – look it up kids! Example, Hz, kHz, MHz, GHz etc.
      http://en.wikipedia.org/wiki/Frequency
      or, http://en.wikipedia.org/wiki/Radio_frequency

      Maxtor DiamondMax 320 GB Buffer size: 16MB Ultra ATA/100 (PATA) P/N: 9DZ0A6-591

      Seq. write: 1024Mb in 33.98s = 30.1MB per second
      Seq. Read: 1024Mb in 14.82s = 69.1MB per second
      Ran. access: 1024Mb in 15.24s = 67.2MB per second

      Flushing cache…
      Sequential read of 1024 chunks of 1024.0kB
      Random read of 8192 chunks of 128.0kB
      Done!

      Remember, the word is THROUGHPUT, or more specifically in this case, bandwidth.

    • Dick

      500Mb in 23.94s = 20.9MB per second
      500Mb in 1.81s = 275.8MB per second
      500Mb in 11.58s = 43.2MB per second
      500mb file 9999 chuncks

      The disk throughput tester claims to clear the cache. It does not, I believe in all cases that show a very high increase in seq reads compared to writes. This can be checked by running HD tune pro in parallel. In my case there is NO DISK read actvity AT ALL, so all reads are from the XP memory cache.

      If I increase the file to 9Gig, the test results in actually reading form the disk (HD tuen) and the numbers of read and writes are not differing to much. (40-60 Mb)

      So the first posting shows only that the desktops cache is cleared and the laptop\’s is not. I also tested my laptop

    • Renzy

      Vista Home Premium
      Toshiba Satellite A200
      Processor- Intel Core2 Duo T7250 2.0 GHz
      Memory- 4GB upgraded from 2GB

      Sequential write – 1936Mb in 106.39s = 18.2 MB Ps
      Sequential read – 1936MB in 77.58s = 25.0MB Ps
      Random access: – 1936MB in 44.99s = 43.0MB Ps

      It seems to slow isn\’t it?? Raymond
      Pleasea reply

      As i am new to this website
      I would like to take this oppurtunity to say, more power to you Raymond. your brilliant!!

      Thanks
      Renzy

    • faraz

      Guys, you are mostly reading cache, not data from the HD – you can’t get speeds like this from a physical drive.

    • Rahul

      Why is random access speed less than sequential access in some cases and greater in some?

    • Corey

      Windows 7
      AMD 3200
      1 gig dual channel pc3200

      write 112.5
      read 104.6
      random 55.7

    • Corey

      Windows XP
      AMD 4400 X2
      2 Gig Dual Channel PC3200

      write 82.6
      read 77.4
      random 35.3

    • RatFink

      In order to accurately measure speed. The file size has to be larger then the RAM you have.

    • ravi

      Thanks for the tool..

      Intel core2 t7200 @ 2.00ghz
      999MHz, 2GB RAM

      Harddisk: Western Digital scorpio 160gb,
      Safeboot encrypted

      windows file cache – disabled.

      375MB in 75.16s = 5.0MB per second
      375MB in 87.25s = 4.3MB per second
      375MB in 142.88s = 2.6MB per second

      Something seriously wrong with my hard disk..
      Any ideas on how to isolate the problem?

    • http://www.jumpbackent.com nFectious

      It locks up on me at 4% complete on the random access test. I gave it about 10 minutes at 4% and it never went anywhere after that. Had the (not responding) message and I had to close down.

    • RMM

      DiskTT FLUSHES the cache after write thus writing all the cached data to the disk. It does not CLEAR the cache after write. So some or maybe all data written to the disk will be in the cache when the read test start. This explains the extreme high read values.

      BTW there is a new version 2 of DiskTT that doesn’t use the cache at all by default. It will still test the thruput of a disk system as experienced by a normal application.
      Try it!

    • RMM

      @ravi
      there is something wrong with your drive (is an IDE or SATA drive?). If it is IDE check your DMA mode under windows.

      or the Safeboot encryption slows your I/O down…

    • Don

      Don’t get too excited with the results yet.

      The laptop 5,400 could be faster than the desktop 7,200 for the following reasons:

      - Has larger cache built in
      - 7,200 RPM drive is on last legs, or greatly reduced
      - Windows set a lower DMA value in Device Manager
      - 2 different windows versions used
      - Need new controller driver for your Sata drive
      - Malware/Spyware/Viruses/Fragmentation affect drive
      - Laptop has fewer startup items (so loads faster)
      - Drive is nearly full on desktop, causing disk thrashing
      - Windows is optimized for speed on laptop

      I am sure there are many other reasons to include on why this is interesting, but definitely NOT a final conclusion.

      Raymond? Might I suggest?

      Get an external drive, update your Sata controllers, & test both drives via high speed USB 3 to see what happens!

    • Ernie

      This is an extreme situation, but I think it’s working. I think I have maxxed out the bandwidth of PCI-E x8:

      Areca ARC-1231 PCI-E x8 card with 4 GB Buffer and 8x Intel 160 GB 34nm SSD drives in RAID 0 configuaration:

      seq write: 1024MB in 1.09s = 936.8MB per second
      seq read: 1024MB in 0.23s = 4371.8 MB per second
      rdm read: 1024MB in 0.23s = 4371.8 MB per second

    • http://alvazan.com Dean Hiller

      very odd, I ran the test and wrote a program that wrote out bytes as well. I only wrote a small file out and got a through put of

      INFO: throughput=177053.82436260625 bytes/second
      INFO: throughput=1416430.59490085 bits/second
      177kB/sec or 1.4Mb/sec (bytes vs. bits as I am comparing to network speeds for an architecture I am working on right now)

      This is much much slower than the results I am told by the other tool….very odd…any ideas why? (The program is in java so I wonder if it is because I am not using native buffers or something to optimize the speed).

      results from tool
      1863MB in 44.48s = 41.9MB/s
      1863MB in 43.12s = 43.2MB/s
      1863MB in 501.81s = 3.7MB/s

      later,
      Dean

    • http://alvazan.com Dean Hiller

      hmmm, I think this tool misreport MB(bytes) and it should be Mb(bits). I ran Iometer and it reports in MB and I got about 5MB which is 40Mb and this tool told me my drive was 40MB which must be Mb actually instead.

      ie. you need to divide by 8 if you want Megabytes instead of it reporting in bits I beleive. This makes more sense and is in line with my other program I wrote then.

    Copyright © 2005-2012 - Raymond.CC Blog