SteveOH

NAS / Software and Hardware RAID Performance Benchmarking

by Steve Hernandez on Feb.15, 2009, under Technology

I recently came across an application provided by Intel for the purpose of benchmarking NAS devices (Define: Network Attached Storage).  I’ve been looking for a solution to benchmark the new RAID-5 array I created on my Windows Server 2008 server (I love the RAID’s redundancy, but the write speeds are quite low).  I wanted to get a better idea of its performance, as Windows will report 250 – 150 MB READS and 20 – 50 MB WRITES, which is considerably poor considering the 4 drives are individually benchmarked at ~100 MB READS and 60 MB WRITES (They are Western Digital Caviar Blues / RE3 / RE16).

The software can be found here or from Intel here.

System Specifications:

Operating System: Microsoft Windows 2008 Enterprise Edition, SP1
CPU Type: Pentium 4 630, 3000 MHz w/ Hyperthreading
Motherboard Name: Dell Dimension 8400
Motherboard Chipset: Intel Alderwood i925XE
Motherboard Frontside Bus Speed: 800 MHz
System Memory (RAM): 4 GB, Dual Channel
System Memory Speed / Timings: DDR2-533 (266 MHz) / 4-4-4-12
Hard Drive Controller Interface: SATA I (1.5 Gbits / s) = 150 MB/s (maximum)
Hard Drives: Western Digital

  • Drive #1 – WDC WD5002ABYS-01B1B (465 GB)
  • Drive #2 – WDC WD5002ABYS-01B1B (465 GB)
  • Drive #3 – WDC WD5000AAKS-00A7B (465 GB)
  • Drive #4 – WDC WD5000AAKS-00A7B (465 GB)
  • Drive #5 – WDC WD800BB-00JHC0 (74 GB) – SYSTEM

Raid Configuration: Software 4 disk RAID-5 (4 x 500GB, 1.36 TB Effective Storage Space)

System Memory Benchmark: Copy – 4790 MB/s, Read – 6266 MB/s, Write – 3341 MB/s, Latency – 92.7 ns
System Processor Benchmark: CPU ZLib – 10592 KB/s, CPU AES  – 3055

Benchmarks completed using Everest Ultimate Edition v4.20.1257

All drives are SATA II (3.0 GB/s) but are running in backwards-capability mode.

The primary roles of the server is to serve files and store backups (File Server) and web and application development and website hosting (Web Server).  I do not stream media, play music from the server or write extremely large amounts of data that are time sensitive.  Therefore, my decision for RAID-5 was based on the space savings (n – 1) storage availability and the redundancy by means of the parity.  I am willing to give up performance (write and read) for storage space and redundancy, but I want to know how much I’m actually giving up.

Tests

  • HD Video Playback, 2x HD Video Playback, 3x HD Video Playback, 4x HD Video Playback

These benchmarks examine the behavior of the NAS unit while (simultaneously) playing one or more HD video files at 720p using Windows Media Player. Intel gives a percentage rate for the sequential reading of data in these tests, which lies at 99.5% for the HD Video Playback Test. With 2x HD Video Playback, it lies at 18.1%. The result is 6.6% with 3x HD Video Playback and 9.6 % with 4x HD Video Playback.

  • HD Video Record

This test writes an HD Video file in 720p format to the NAS unit. This test is made of up of mostly sequentially transferred data.

  • HD Video Playback & Recording

HD Video Play & Record examines the behavior of the NAS unit when simultaneously reading and writing an HD Video file in the 720p format. The sequentially-transferred data in this test is approximately 18% of the test.

  • HD 2x Playback 2x Record

This benchmark is similar to the one above, but the proportion of sequential file operations is 3%.

  • HD Playback With Office

This metric measures the data transfer rates when an HD Video file is read from the NAS unit while working with the Office applications. This test is made up of 608 files. The proportion of sequential file operations is 53.2%.

  • HD Playback With Backup

Like the previous test, but this time an HD Video file is played while simultaneously carrying out a backup on the NAS unit.

  • Content Creation

This benchmark is made up of 95% write operations to the NAS unit. This simulates the creation of files on the NAS unit such as is the case when, for example, using video editing programs.

  • File Copy To NAS / File Copy From NAS

These tests determine the data transfer rate when copying files to or from the NAS unit. In both of these test processes, a 4 GB file is copied. Unlike with Backup / Restore, 64 KB is read and written.

  • Directory Copy From NAS / Directory Copy To NAS

Similar to the previous test, files are copied to and from the NAS unit. A total of 126 files with a total size of approximately 188 MB are written and read across the network.

  • Photo Album

This test determines how the NAS unit handles the supply of a multitude of small files—for example, viewing digital photographs stored on the NAS unit. It simulates the viewing of a total of 169 photographs with an overall size of approximately 1.2 GB.

Considerations and Assumptions

These tests were conducted on the Server, not over the network, which will probably result in very high throughput.  I did this to get a feel for the actual performance, not that felt over the wire, and because I’m currently running 100 Mbps (Cat 5e) on my LAN, not gigabit.  Thus, I did not want the network to be the bottleneck in these tests, as it would be saturated and skew the results.

I ran the tests 6 times, in each, manipulating the environment to establish a real world benchmark.  Since this benchmark is done via an application, it will compete with other processes for resources.  Additionally, it will compete with the actual Operating System who is responsible for transfering the data to the disk and calculating the parity for the raid (which is quite memory and CPU intensive).  For test 6 I ran the system in Safe Mode, utilizing only necessary processes and services to run the operating system (Safe Mode without Networking).  Thus, the numbers are quite high.  In the others, I ran Outlook, SQL, browsed the internet and downloading files via torrents (to the actual RAID, so this will mess with the numbers as well) – not all at the same time, of course, except for test 4.

Results

Test 1 2 3 4 5 6 Avg MB/s
HDVideo_1Play 267.789 267.789 80.364 54.54 59.134 274.603 167.37
HDVideo_2Play 223.736 223.736 68.638 44.619 75.123 223.204 143.18
HDVideo_4Play 177.201 177.201 70.222 54.731 73.009 180.005 122.06
HDVideo_1Record 626.098 626.098 651.41 618.888 678.788 732.636 655.65
HDVideo_1Play_1Record 82.744 82.744 85.224 71.802 79.058 155.986 92.93
ContentCreation 6.199 6.199 5.983 5.628 6.141 6.022 6.03
OfficeProductivity 25.383 25.383 25.474 20.958 26.249 27.955 25.23
FileCopyToNAS 633.475 633.475 748.91 690.827 768.598 806.578 713.64
FileCopyFromNAS 147.096 147.096 79.675 57.42 80.561 151.573 110.57
DirectoryCopyToNAS 65.156 65.156 25.13 18.384 23.418 81.739 46.50
DirectoryCopyFromNAS 44.048 44.048 20.412 15.53 20.929 47.274 32.04
PhotoAlbum 27.748 27.748 26.411 19.508 28.171 29.204 26.47

Discussion

I attribute the absurd file copy speeds to caching and buffering at both the OS, RAID and the Drive level, as well as the writes for the video recording.  All in all, I’m pleased with the benchmarks and am satisfied with the set up.  One thing to note is that the tests were done from the SYSTEM drive to the RAID, and thus, the SYSTEM drive introduced an additional bottleneck as the transfers from it to the RAID were going to be quite slow (the drive is benchmarked at ~56MB READ and ~30 WRITE).

:, , , , , , , , , , , , , ,

Leave a Reply

Looking for something?

Use the form below to search the site:

Still not finding what you're looking for? Drop a comment on a post or contact us so we can take care of it!

Visit our friends!

A few highly recommended friends...