1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems

Facebook RSS Twitter Twitter Google Plus

Phoronix Test Suite


Mobile HDD Cache Comparison

Michael Larabel

Published on 18 December 2005
Written by Michael Larabel
Page 1 of 4 - Comment On This Article

When it comes to standard 2.5" hard drives designed for mobile operation, 4200 and 5400RPM generally reign for notebooks while 7200RPM HDDs are available for those attempting to achieve desktop-like performance when on the road. On the contrary, the spindle RPM increase does come at a hefty cost when it comes to the noise emitted, heat generated, and the longevity of the notebook battery. Although these disadvantages may be acceptable for some, others would prefer to be able to keep a quiet notebook on their lap without the need for any after-market cooler such as the Prime Cooler CoolPad or Vantec's LapCool. Others of course would simply like to simply spend the afternoon with their mobile device without being tied against the wall or an alternative power source. Similar to desktop HDDs, one of the ways to generally increase the computing performance for a notebook without purchasing a greater RPM drive is through bumping up the onboard cache size from the common 8MB, or even 2MB, capacity to the level of 16MB. For the uninitiated, when a call is made to the hard disk drive the cache is first searched to see if any segments are present before fetching the data from the disk platters. Thus since the HDD cache size is so incredibly small compared to the actual capacity of the present day drives, a greater amount can often be beneficial especially when recursive tasks are executed, which of course is similar to increasing the overall system memory. For today, we have prepared some benchmarks comparing the performance numbers of an 8MB ATA-6 2.5" notebook hard drive against that of a similarly equipped 16MB cache buffer HDD. This testing had been carried out to see just how well the additional cache affects the overall system performance in addition to the HDD centric penguin benchmarks.

The two contenders for our 8MB versus 16MB HDD cache comparison are Hitachi's Travelstar HTS541080G9AT00 and Toshiba's Slim MK1032GAX. Although this precisely is not a controlled environment due to slight variations in the logic circuitry and other hardware components their specifications are near identical when it comes down to the 5400RPM, ATA-6 interface, 5.5ms average latency, and 12ms average seek time. Of course, the Toshiba 100GB drive offers a 16MB cache while the Hitachi 80GB sticks with the normal 8MB capacity. For testing these two HDD devices, we turned to our faithful IBM/Lenovo R52 notebook.

Hardware Components
Processor: Intel Pentium M 750 (1.86GHz)
Motherboard: IBM R52 18494WU (i915PM + ICH-6M)
Memory: 1 x 512MB DDR2
Graphics Card: ATI RADEON X300 64MB (dedicated)
Optical Drives: DVD-RW Drive
Add-On Devices: Intel PRO/Wireless 2915ABG
Software Components
Operating System: FedoraCore4
Linux Kernel: 2.6.14-1.1653_FC4
GCC (GNU Compiler): 4.0.0
Graphics Driver: ATI v8.20.8
Xorg: 6.8.2

Having their hand at comparing the two hard disk drives is hdparm, tar, Iozone, and Disk Wriggler. With hdparm we used the -t flag to perform timed disk reads onto the hard disk while we turned tar into a real-world benchmark using the Linux time command and specifying to record the amount of time to archive the Unreal Tournament 2004 v3334 demo, and then the amount of time needed to extract the same file. With Iozone, we ran this dedicated multi-OS hard drive benchmark with various parameters but consisting of read/writes at 100MB and 500MB. The next HDD benchmark, which is its first time making an appearance at Phoronix is Disk Wriggler. The purpose of the benchmark, which is created by the Australian based XDT and released under the GNU GPL, is to test the file system storage throughput of video files. With Disk Wriggler, we used the NTSC, HD, and 4K frame type standards. Where NTSC is engineered for 720 x 486 8-bit 4:2:2, 4K is the maximum frame type in Disk Wriggler with parameters of 4096 x 3112 10-bit 4:4:4. The HD frane type on the other hand uses a resolution of 1920 x 1080 10-bit 4:4:4. All in all these Linux benchmarks should be able to do an appropriate job reflecting the performance level between the two hard drives and primarily the difference in onboard cache. Let us now jump right into the benchmarks!

Latest Linux Hardware Reviews
  1. CompuLab Intense-PC2: An Excellent, Fanless, Mini PC Powered By Intel's i7 Haswell
  2. From The Atom 330 To Haswell ULT: Intel Linux Performance Benchmarks
  3. AMD Radeon R9 285 Tonga Performance On Linux
  4. Apotop Wi-Copy
Latest Linux Articles
  1. AMD Moves Forward With Unified Linux Driver Strategy, New Kernel Driver
  2. MSI: Update Your BIOS From The Linux Desktop
  3. NVIDIA vs. AMD 2D Linux Drivers: Catalyst Is Getting Quite Good At 2D
  4. 15-Way GPU Comparison With Mesa 10.3 + Linux 3.17
Latest Linux News
  1. Phoronix Test Suite 5.4 M3 Is Another Hearty Update
  2. GParted 0.20 Improves Btrfs Support
  3. EXT4 In Linux 3.18 Has Clean-ups, Bug Fixes
  4. Emacs 24.4 Has Built-In Web Browser, Improved Multi-Monitor Support
  5. NVIDIA's NVPTX Support For GCC Is Close To Being Merged
  6. KDE's KWin On Wayland Begins Using Libinput
  7. Khronos Releases OpenVX 1.0 Specification
  8. Linux Kernel Working Towards GNU11/C11 Compatibility
  9. Ubuntu 15.04 Is Codenamed After A Monkey: Vivid Vervet
  10. Following GCC, Clang Looks To Default To C11
Latest Forum Discussions
  1. HOPE: The Ease Of Python With The Speed Of C++
  2. Users/Developers Threatening Fork Of Debian GNU/Linux
  3. Updated and Optimized Ubuntu Free Graphics Drivers
  4. Bye bye BSD, Hello Linux: A Sys Admin's Story
  5. NVIDIA Presents Its Driver Plans To Support Mir/Wayland & KMS On Linux
  6. AMD Is Restructuring Again, Losing 7% Of Employees
  7. Open-Source AMD Fusion E-350 Support Takes A Dive
  8. Upgrade to Kaveri, very slow VDPAU performance