Contact Us | Careers
 

Print

Packages are too inflexible and don't provide competitive differentiation, but custom-built solutions are too slow [to develop] and costly. System integrators are starting to offer next-generation pre-built solutions. - Gartner Research

 Did you know?

xSynthesis Search includes features often found only in enterprise search packages, at the cost and ease of use of entry-level search engines. Use it for your publicly accessible e-commerce, membership, or marketing web site, your protected members-only content, or your intranet. 

 

 Interface Speeds

Here are some theoretic interface speeds:

  • PCI: 133 MB/s
  • USB 2.0: 60 MB/s (480 Mbps)
  • USB 1.1: 1.5 MB/s (12 Mbps)
  • USB 3.0: 640 MB/s (5 Gbps)
  • SATA 1: 150 MB/s
  • SATA 2: 300 MB/s
  • SATA/600: 600 MB/s
  • PCIe 1.1: 250 MB/s per channel
  • PCIe 1.1 1x: 250 MB/s
  • PCIe 1.1 8x: 2,000 MB/s
  • PCIe 2.0: 500 MB/s per channel (note: usually 2.0 cards are compatible with 1.1 motherboards)
  • AGP 1x: 250 MB/s
  • 1 Gigabit network/ NIC: 128 MB/s (1,024 Mbps)
  • 100 Mbps network/ NIC: 12.5 MB/s
  • FireWire 800 (80 MB/s)

For more on interface speeds see this article.

 

 

very Fast Hard Drive Performance (Above 500 MB/s) On A Budget (Using 3.5" HDDs, 2.5" HDDS, SSDs, and RAID 0)

by Zoran Zaev, 2009-03 (March) to 2010-01-14 (January)

In this article, I review various hard drives (HDDs) and various hard drive configurations, particularly RAID 0 configurations for attaining top performance and data transfer speeds with inexpensive hardware. Furthermore, some drive controllers will be reviewed and compared.

You may be thinking how you can improve the performance of your own desktop, workstation, or small business server? Or, you may be considering how you can improve the performance of your current hard drives. You may be surprised that you can improve performance without spending a fortune.

First, you may like to review what hard drives and drive controllers you currently have. List the drives, make, model number, and the same for the drive controller. Next, you can search via Google or others search sites and find the specification details of those particular drives, such as some of the more important being interface type (IDE, SATA), rotation speed (5,400 rpm, 7,200 rpm, 10,000 rpm, etc), buffer size (8 MB, 16 MB, 32 MB, etc).

Test Setup

I will be testing various hard drives, from 3.5" to 2.5" types, and Solid State Drives (SSDs). I will be testing them using the following main computers & controllers:

  1. Desktop PC 1: FoxConn G43MX-K Motherboard onboard SATA Controllers (1 @ 4 x SATA connectors, 1 @ 2 x SATA connectors), rated at 300 MB/s data transfer support;
    - PPA SATA 3 Port Model 1356/ Sil 3132, PCI based Sata Card, 300 MB/s, 2 independent channels card (ex. at NewEgg, TigerDirect, at manufacturer PPA-USA)
    - Rosewill 4-port, RC209EX, Sil 3114, PCIe 1x based Sata Card, upto 150MB/s, 4-channel RAID adapter (at NewEgg, the manufacturer Rosewill)
  2. Desktop PC 2: Biostar G31 Motherboard with onboard SATA Controller (1 @ 4 x SATA connectors)
  3. Notebook 1: HP dv7-3060us with onboard SATA controller (1 @ 2 x SATA connectors)
    - USB 2.0 controllers on the HP dv7 Notebook and on the FoxConn workstation PC
    - Express Card, Rosewill RC-605, 2 port eSATA II card, upto 300 MB/s (see NewEgg, the manufacturer Rosewill)
  4. Notebook 2: Acer 5520 with onboard SATA controller (1x)
  5. Notebook 3: Dell Inspiron 5150
  6. Netbook: HP Mini 1116NR with onboard ATA/ IDE controller
    - USB 1.1 controller on the HP Mini 1116NR

Benchmark Results

The benchmarking was primarily done using ATTO Disk Benchmark v.2.34 (older, but very useful utility, for download search the web or go to TechPowerUp). Also, you may like to take a look at HD Tune, which can show the drive temperature among other things, but it cannot measure the speed of software RAID 0 configurations, which I will use in this article. An overall very helpful testing program, not just for hard drives is PassMark Software's PerformanceTest (they also have great online statistics: CPU, Video, Hard Drive, and System benchmarks to compare with other users). However, PerformanceTest drive testing capability is weak.

1. Single Drive Benchmarks

The following is a selection of benchmarks for the read and write speeds for various hard drives. Please note that small file reads/ writes (or throughput) are typically more important for applications that deal with small files often and large files read and write (or throughput) impacts applications that deal with large files such as graphics, video, as well as Virtual PC operation. I decided to focus on throughput and not review some other statistics such as access time, burst rate, or CPU usage. Depending on your needs, you may like to decide to review and benchmark for those as well, but I would suggest that throughput is the most important statistic impacting performance
 

System/ Configuration,
Drive Specification (MB stands for "Motherboard")
Small files (0.5kb) [1] Large files
(8,192kb/8mb) [1]
PATA Drives, 2.5" Size PC/ Controller Read Write Read Write
Fujitsu PATA MHS2030AT 2.5", 30 GB, 4,200 rpm, ATA-100 (transfer speed 19.4 to 38.1 per spec), 2MB buffer (spec, see page 24) 2.5 USB 2.0, Dell Dimension 8100 0.9 0.9 17 15
Fujitsu PATA MHV2080AH 2.5", 80GB, 5,400 rpm, 8MB cache (max transfer speed 55.4 MB/s), 4MB buffer, (spec, see page 24) 2.5 MB/Tablet ? ? 34 ?
Western Digital WD1600BEVE-00UYT0, PATA WD160G Scorpio, 2.5", 160GB, 5,400 rpm, 8MB buffer (spec) 2.5 Inspiron 5150/ IDE on MB 3.8 4.1 48
(best!)
21
SanDisk pSSD PATA 16GB on ICH7 UATA 82801G Intel Connector, 2.5", SSD (spec) 2.5 On HP Mini 1116NR 1.9 0.6 38 17
PATA Drives, 3.5" Size PC/ Controller Read Write Read Write
Maxtor DiamondMax Plus 40, 54098U8, PATA, 3.5", 40 GB, Ultra 66 3.5 MB/ Dell Dimension 6.6 3.6 26 25
Samsung SP0802N, 80 GB, PATA, 3.5", buffer 2 MB, UDMA 100, 7,200 rpm (spec) 3.5 MB/ Dell Dimension 0.5 0.5 46
(best!)
48
(best!)
SATA Drives, 2.5" Size PC/ Controller Read Write Read Write
Hitachi Travelstar HTS542512K95A00, 2.5", 120GB, TravelStar, 5K250, 5400 rpm, 8MB, 9.5ms, SATA 2.5 Acer 5520/ SATA on Enclosure 4.0 2.9 51 51
Seagate ST9500420AS, 500MB, 2.5", SATA (spec) 2.5 MB/ HP dv7-3060us 6.0 5.3 97
(best!)
95
(best!)
SATA Drives, 2.5" Size PC/ Controller Read Write Read Write
Maxtor 7Y250M0, 250MB, 3.5", 7,200 rpm, SATA 150, 8MB cache 3.5 Biostar G31 5.8 4.4 56 58
Maxtor 6B300S0, 300GB, 7,200 rpm, 3.5", SATA 150, 16 MB cache 3.5 Biostar G31 12.0
(best!)
11.0
(best!)
50 56
Hitachi DHT725050VLA 380, 500 MB, 3.5", SATA, 8MB buffer, UDMA 6 3.5 MB ? ? 74 ?
Western Digital WD150 Raptor 10,000rpm, 150GB, WD1500AHFD UDMA 6 (ATA/133) 3.5 FoxConn G43MX-K 12.0
(best!)
10.0
(best!)
65 74
Western Digital WD150 Raptor 10,000rpm, 150GB, WD1500AHFD UDMA 6 (ATA/133) 3.5 Biostar G31 9.6 9.1 83 83
Western Digital Caviar Green WDCWD10EACS, 1TB, 32MB (or 16 MB?), 5400 rpm+ 3.5 FoxConn G43MX-K 4.0 5.4 76 77
Samsung SpinPoint F1, HD103UJ, 1TB, 7,200 rpm, 32MB cache, UDMA 7 (ATA/512) 3.5 FoxConn G43MX-K 4.9 3.9 92 84
Seagate ST3150034 IAS, 7200 rpm, 32MB, 1.5TB 3.5 FoxConn G43MX-K 9.9 8.4 107
(best!)
106
(best!)
Network Drives/ NAS Size PC/ Controller Read Write Read Write
Network (NIC) drive at 100 Mbps
Note: helping when considering Network Attached Storage (NAS)
3.5 NIC 100 Mbps, 12.5 MB/s (100 Mbps) 2.3 1.2 11 11
NAS at 1000 Mbps (gigabit) 3.5 NIC 1 Gbps, 128 MB/s 2.6 1.3 44 38
NAS at 1000 Mbps (gigabit), after fresh reboot of Windows 7 3.5   2.7 1.4 59 48
NAS at 1000 Mbps (gigabit), turned off client firewall 3.5   2.9 1.4 79 58
NAS at 100 Mbps, via WebDAV 3.5 NIC 100 Mbps, 12.5 MB/s 1.0 0.3 97 93
Special: Virtual PC Drives, Spanned Volume Size PC/ Controller Read Write Read Write
Virtual PC 2007 running on Western Digital WD150 Raptor     3.2 2.3 42 58
Virtual PC 2007 running on Raid0_4dr     3.1 1.6 63 75
Spanned Volume: 4dr, Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x2)   4 drives SPANNED volume 5.9 7.4 75 76

Notable observations:

  1. PATA drives seem to top around 38-48 MB/s rate, though I would have expected them to do better. I'm not sure if the interface speed or the specific drives were the bottleneck. Further research and testing of some other drives could help bring an answer in that respect. Some of the following ought to do better, with expected reads/ writes in ~ 60 MB/s range, though you have to watch to make sure that the IDE controller you have is not older and becomes the bottleneck (in these charts no PATA drive appears to give throughput higher than 36 MB/s).
    1. Transcend TS16GSSD25-S 2.5" 16GB PATA SLC Internal Solid State Drive (SSD), $169; spec claims 74 read/ 62 write capability (at NewEgg)
    2. Western Digital Scorpio Blue WD1200BEVE 120GB 5400 RPM 8MB Cache 2.5" ATA-6 Internal Notebook Hard Drive, $74.99 (at NewEgg)
    3. Western Digital Scorpio Blue WD800BEVE 80GB 5400 RPM 8MB Cache 2.5" ATA-6 Notebook Hard Drive, $59.99 (at NewEgg)
    4. SAMSUNG Spinpoint M Series HM080GC 80GB 5400 RPM 8MB Cache 2.5" ATA-6 Notebook Hard Drive, $69.99 (at NewEgg)
    5. 5400 RPM & Speed: Hitachi's Travelstar 5K250 and WD's Scorpio WD3200BEVT
    6. 160 GB Notebook Drive Roundup
  2. While the famous WD Raptor does very well with small files, it does not do as well with large files. This means that it is ideal for being the OS drive, for fast Windows boot up, but for handling large files, such as graphics, video files, etc. drives like the Seagate ST3150034 certainly does great. The Seagate ST9500420AS 2.5" has an amazing performance, too.
  3. "Green" drives are noticeably slower than regular drives for the same price and I would not recommend them, unless you are using them for storing rarely accessible data, such as backup data. They could make fine external backup drives, however, for hot-backup purposes.
  4. NAS at 1000 Mbps/ 1 Gbps was surprisingly low ( ~40 MB/s). There has to be a certain bottleneck for this to be taking place. Perhaps a firewall interference? The networked drive being accessed is a RAID 0 with 4 drives, performing at a very high speed. Another impact may be the network switch.
  5. It was surprising that NAS 100 Mbps, operating via WebDAV (this means you can map a regular Windows drive over HTTP) performed very poorly for small files (perhaps due to the WebDAV overhead), but very fast for large files (~ 95 MB/s) and that was surprising (perhaps I have a gigabit card on that remote server, without realizing it?)
  6. The impact of the firewall on NAS at 1000 Mbps is very interesting, but it seems to not impact WebDAV communications. Further research into firewall settings would be worth while. Brief testing appears to show that running a firewall such as TrendMicro on the server reduces the network throughput by about 10 MB/s for mid to large files, no impact for smaller files. Running a client side firewall slows the throughput another 5-10 MB/s. You may like to read the following network optimization articles:
    1. Network performance and data throughput may be significantly slower after installing Windows XP Service Pack 2
      http://support.microsoft.com/?kbid=842264
    2. Squeeze Your Gigabit NIC for Top Performance by Charlie Schluting at Enterprise Networking Planet
      He says that "For attaining maximum throughput, the most important options involve TCP window sizes" and he gives instructions how to do this.
  7. When considering NAS Storage, do compare it to USB and eSATA. Unless you need it to be NAS, USB 2.0 may even do just fine.

2. Drive Controller/ Enclosure Impact Benchmarks

Now, I took some very fast SSD drives and tested them with different drive controllers, even by putting them into different drive enclosures to see what the performance difference may be. The results were surprising.

System/ Configuration,
Drive Specification
Small files (0.5kb) [1] Large files
(8,192kb/8mb) [1]
Drive PC/ Controller Read Write Read Write
OCZ Vertex Series OCZSSD2-1VTX30GXXX 2.5" 30GB SATA II MLC SSD (see at NewEgg) Via PPA SATA 3 Port Model 1356/ Sil 3132, PCI Card 10.1 3.6 97 63
OCZ Vertex Series OCZSSD2 Via Rosewill 4-port, RC209EX, Sil 3114, PCIe 1x Card 11.6 4.2 133 126
OCZ Vertex Series OCZSSD2 Via PCIx/ Express Card Rosewill RC-605 12.0
(best!)
6.8
(best!)
133 95
OCZ Vertex Series OCZSSD2 Via external enclosure Patriot Dual Bay 2.5", on SATA port, 3.5" drive bay PC25S (see at Amazon, TigerDirect, at NewEgg) 11.3 3.5 140 125
OCZ Vertex Series OCZSSD2 Via 2.5" external enclosure on eSATA port, Rosewill (very similar to this one at NewEgg, USB 2.0 & eSATA) 10.7 3.6 173 154
OCZ Vertex Series OCZSSD2 On HP dv7-3060us, onboard eSata port, direct connection, no enclosure. 11.0 3.4 250
(best!)
163
(best!)
Patriot Warp 32 GB, PE32GS25SSDR, SSD, 2.5" (see at NewEgg) Acer 5520/ SATA on MB 2.8 4.2 153
(best!)
94
(best!)
Patriot Warp 32 GB, SSD, 2.5" Via Rosewill 2.5" enclosure 2.2 3.2 129 91
Patriot Warp 32 GB, SSD, 2.5" Via PCIx card and Rosewill 2.5" enclosure 2.9 4.4 116 71
Patriot Warp 32 GB, SSD, 2.5" Via USB 1.1 and Rosewill 2.5" enclosure (possibly USB 2.0, though) 0.8 1.0 31 24

Notable observations:

  1. The PCI card appears to be limited to around 100 MB/s per drive, perhaps by the PCI interface itself (133 MB/s theoretic speed limit).
  2. The PCIe 1x card appears to be limited to 133 MB/s reading speed, though PCIe 1x theoretic speed limit is 250 MB/s. The bottleneck here is likely that the card is rated for Sata 1, 150 MB/s drives.
  3. Similar situation is occurring with the PCIx card, though this is surprising because this card is rated at Sata 2, 300 MB/s.
  4. It was very surprising that the enclosures impacted the performance of the drive, slowing it down to 140 MB/s or 173 MB/s, far below its limit of 250 MB/s! The enclosure circuitry does appear to have impact!
  5. Why did the PCIx card excel in writing small files? Perhaps something within its circuitry or interface? A cache that can batch these small write requests better? Not sure.
  6. Not clear why the Patriot Warp SSD drive over USB 1.1 would go up to 31 MB/s? The only explanation is that particular USB on the HP Mini is actually USB 2.0, but it cannot really bring up to 60 MB/s throughput due to the USB architecture.
  7. The Patriot Warp 32 GB SSD shows a slow down when using e-SATA enclosure vs connecting directly the motherboard. An additional slow down occurs if this drive goes via a PCIx card. Theoretic speed for this drive is: 175 read/ 100 write MB/s and the actual best reading seems to be very close to it as we would expect.
  8. We can see that SSD drives excel, particularly in reading large files, though some of the newer ones such as the OCZ Vertex really outperform almost in any category! And, their prices have come down to around ~$100/ drive.
  9. Some controllers and even operating systems are not optimized for using SSD drives. A program like this may help: Managed Flash Technology (MFT) Flash SSD Acceleration Software.

3. Fast RAID 0 Benchmarks

What can these typical, consumer-level drives do when put in Raid 0 (stripping only) configuration do in terms of performance? It turns out that we can get some amazing performance benefits, just by using Windows XP/ Vista/ 7, etc. built-in software Raid 0 (stripping) capability. Here are some of the results.

System/ Configuration,
Drive Specification
Small files (0.5kb) [1] Large files
(8,192kb/8mb) [1]
Drive PC/ Controller Read Write Read Write
Raid0 (Sw)= Sam 1TBx2, WD 1TBx1   4.5 5.3 231 224
Raid0 (Sw)= Sam 1TBx2   5.2 6.4 223 147
Raid0 (Sw)= Sam 1TBx2, Hw Raid0 - 500MBx2, WD 1TB   5.6 6.0 269 274
Raid0 (Hw)= Hitx2   10.2 4.8 109 94
2 Drive Raid 0 PC/ Controller Read Write Read Write
Hitachi DHT725050VLA (x2) HW HW Raid 0 on PPA SATA 3 Port Model 1356/ Sil 3132 PCI Card 10.2 4.8 109 94
2 Drv, OCZ SSD (x1), Patriot SSD (x1) 1xMB, 1xPCIe 3.8 5.5 179 74
2 Drv, OCZ SSD (x1), Patriot SSD (x1) 2xPCI 5.1 6.7 136 74
2 Drv, OCZ SSD (x1), Patriot SSD (x1) 2xPCIe 4.5 6.8 132 68
2 Drv, Patriot SSD (x1), Hitachi Travelstar HTS542512K (1x) Hitachi on External Enclosure, SSD on MB 3.4 3.6 64 64
2 Drv, Maxtor 7Y250M0(x1), Maxtor 6B300S0 (x1) on BiostarG31 MB 8.2 7.2 118
(best!)
120
(best!)
2 Drv, Samsung SpinPoint F1 (x2) On MB 4.9 7.9 213
(best!)
214
(best!)
2 Drv, Seagate ST9500420AS (x2) MB, PCI 12.0 4.2 150 147
2 Drv, Seagate ST9500420AS (x2) MB, PCIx 8.3 8.1 145 145
2 Drv, Seagate ST9500420AS (x2) MB 5.9 8.1 150 155
2 Drv, Seagate ST9500420AS (x1), Samsung SpinPoint F1 (x1) MB 8.5 4.2 154 147
3 Drive Raid 0 PC/ Controller Read Write Read Write
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x1) MB 5.4 6.7 290
(best!)
294
(best!)
WD Caviar Green (1x), Hitachi DHT725050VLA (x2) HW MB, Hitachi in hardware RAID 0 on PCI card 6.0 5.4 149 149
Samsung SpinPoint F1 (x2), WD Caviar Green (1x) On MB 4.5 5.3 231 224
4 Drive Raid 0 PC/ Controller Read Write Read Write
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x2) MB 6.0 7.0 431
(best!)
378
(best!)
Seagate ST9500420AS (x2), Hitachi DHT725050VLA (x2) HW MB (Seagate), PCI hardware RAID (Hitachi) 5.8 4.9 288 252
Samsung SpinPoint F1 (x2), Hitachi DHT725050VLA (x2) HW MB (Samsung), PCI hardware RAID (Hitachi) 5.9 4.7 284 255
5 Drive Raid 0 PC/ Controller Read Write Read Write
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x3) MB 6.3 7.9 488
(best!)
353
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x3) MB, PCIe 4.6 6.3 432 330
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x3) MB, one ST on PCIe 3.5 6.3 556
(best!)
392
(best!)
Samsung SpinPoint F1 (x2), WD Caviar Green (1x), Hitachi DHT725050VLA (x2) HW MB, Hitachi on PCI as hardware RAID 0 5.6 6.0 269 274
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x2), Patriot SSD (x1) MB, SSD on PCI 4.1 5.9 390 272
6 Drive Raid 0 PC/ Controller Read Write Read Write
Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x2), Hitachi DHT725050VLA (x2) HW MB, Hitachi on PCI hardware RAID 0 6.1 6.3 491
(best!)
346
Samsung SpinPoint F1 (x2), Hitachi DHT725050VLA (x2) HW, Seagate ST9500420AS (x1), WD Caviar Green (1x) MB, Hitachi on PCI hardware RAID 0 6.1 5.8 348 312
7 Drive Raid 0 PC/ Controller Read Write Read Write
Samsung SpinPoint F1 (x2), Hitachi DHT725050VLA (x2) HW, Seagate ST9500420AS (x2), WD Caviar Green (1x) MB, Hitachi on PCI hardware RAID 0, some on PCI or PCIe x1 cards (MB is limited to 6 - 1 OS drive = 5 drives) 5.8 5.5 370 341

Notable observations:

  1. The OCZ/Patriot SSD (2 drv) configuration is useless, because of the limitation of the controllers, plus the speed seems to drop to the slower of the two hard-drives (Patriot SSD), minus the controller slow down. It is actually faster to use the OCZ drive by itself, connected directly to the MB!
  2. It is unusual that the 2 Drv RAID 0 of Samsung drives perform so well, and the 2 Drv RAID 0 Seagate set perform more poorly, even though the Seagate drives are faster a bit when running by themselves?
  3. Note that adding one slow drive (WD Caviar Green) to 2drv fast array (Samsung SpinPoint F1 (x2)) adds only 20 MB/s added speed, certainly not worth doing. Adding the same WD Caviar Green to the Hitachi hardware array, also only added 40 MB/s extra speed.
  4. The best performance, given the usage of drives appears to be Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x2) (4 drives) and Samsung SpinPoint F1 (x2), Seagate ST9500420AS (x3) (5 drives), going up to 431 MB/s and 556 MB/s reading throughput! This is great for using only 4 to 5 drives. Adding slower drives to these, actually reduced the overall speed. Even adding that 5th drive didn't help that match, likely because we reached the maximum speed that my drive controller can handle on the FoxConn G43MX-K motherboard. Faster controller would be need to do better with more drives.
  5. For further research:
    1. RAID Stripe Size Discussion, here's the key point/ quote:
      "Conventional hard drives deliver their best transfer performance when they read or write sequentially, repositioning the heads as little as possible. From this standpoint, it makes the most sense to select the largest stripe size available, especially if your hard drives are good at providing high throughput. However, this only works if the files stored or read are at least as large as an entire stripe. If you will end of storing millions of text files, Word documents, small spreadsheets or similar small files, small stripe sizes will help to distribute all files across multiple drives to keep throughput high."
      From RAID Scaling Charts, Part 3: 4-128 kB Stripes Compared at Tom's Hardware
      In the benchmark, it appears that 32KB and 64KB stripes are best.
    2. RAID Scaling Charts, Part 1, at Tom's Hardware
      http://www.tomshardware.com/reviews/RAID-SCALING-CHARTS,1635.html
      They had a similar bottleneck after 5 drives in RAID 0 and thought that the bottleneck must be the controller or the motherboard. The controller limited their write speeds to 275 MB/s.
    3. RAID Scaling Charts, Part 2, at Tom's Hardware
      http://www.tomshardware.com/reviews/RAID-SCALING-CHARTS,1662.html
    4. Picking the right stripe size
      http://storageadvisors.adaptec.com/2006/06/05/picking-the-right-stripe-size/

 

Conclusion

I hope some of the information in this article was helpful in your work on improving the performance of your hard drive subsystem. In the next article, I will review some further hard drive and controller configuration ideas.

 

Resources

[1] Patriot 128GB Warp SSD Drive PE128GS25SSDR, Benchmarking Review

[2] A Brief History of the Hard Disk Drive

 

Want to know more? today and let us know how we can make this section better.

| Home | Products | Services | Resources | Our Guarantee | About Us | Contact Us |

Copyright © xSynthesis 2001-2013. All Rights Reserved. Terms of Use and Privacy Policy.
484 Williamsport Pike #126, Martinsburg, WV 25404.

  

Page details:
Page name: hw_hard_drive_performance.aspx
Path: /en/resources/hw_hard_drive_performance.aspx