Which drive to build a matrix

Rate this item
(0 votes)

RAID should be built from disks with the same parameters. It's not just about volume, but also the average reading, time of access, cache or speed. We already know how it should be, but what if at home we have an older drive that you want to plug into our matrix? Such a scenario is not difficult to imagine. I decided to test this configuration to see how it will be checked in practice.

For these purposes I have used a little worn-00MJB0 WD2500KS disk capacity of 250GB. This CD was chosen for a reason. As we recall capacity of RAID 0 is a multiple of the smallest drive times the number of disks. Since that my mother was made of two discs with a capacity of 250GB adding a third disc of the same capacity of the matrix A 750GB in size, thus making full use of available free space on all three drives.

Data Retrieval in Boston - raid data recovery and critical server data salvaging services guide.

HD Tach – Long - WD2500KS-00MJB0:

HD Tach – Long - WD2500KS-00MJB0 test

As shown in the screenshot above disk parameters are not the best, today's standards, its average reading is only 55.5 MB / s, which compared to the ST3250410AS characterized by reading at 78-80MB / s, of which I have constructed a matrix, goes quite pale. After all, it's time to see how adding this disk will affect the performance of the presented array.

The tests were divided into two parts. First, we compare the matrix formed by the two drives to an array constructed with three disks and with "Write-Back Cache", while the second part of the same matrices compare now with the option "Write-Back Cache". Tests were conducted in the three most popular programs, disk benchmarks like HD Tach, HD Tune and Everest. Time for testing.

Part I. Tests with the option "Write-Back Cache" off.

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF:

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF test

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF test 2

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF result

Analyzing the results of this test shows that the average reading has increased by only 3MB / s, while the same graph in the case of three disk array is very "jagged", to blame our extra hard, which significantly deviates from the other disk parameters included in the matrix.

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF:

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF test

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF test 2

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF result

A similar situation occurs in the HD Tune without enabled "Write-Back Cache" - here the average reading has increased by more than 6MB / s and both the same graph, as well as the access time field are significantly more scattered.

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF:

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OF test

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OF test 2

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB OFF result

Everest also, as well as previous benchmarks, showed only a slight increase in speed compared to two disk array configuration, only the access time decreased in a fairly significant degree.

Worth to know what you can do to raid array repair.

Part II. Tests with the option "Write-Back Cache" on.

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON:

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test 2

HD Tach – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON result

When you click on "Write-Back Cache" graph is already "calmed down", while the average reading is less than two disk array in the matrix! This is no great difference, but it is, you can see here exactly that the addition to the slower disk arrays slowed it all.

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON:

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test 2

HD Tune – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON result

A similar situation in the HD Tune results here to read the two matrices almost identical, while the access time very scattered.

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON:

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON test 2

Everest – Long – 2x ST3250410AS vs. 2x ST3250410AS + WD2500KS-00MJB0 WB ON result

In the built-in Everest benchmark one can see that our three disk array takes down the two disk array losses everywhere except at the beginning of the disk read and access time.

After a thorough analysis of all tests can be concluded that the coupling of the disk array significantly different parameters is not the best option. In this case, productivity growth is small, and the same "work culture" of the matrix strongly disturbed, when we add more tests with the option "Write-Back Cache", we see that the three-disks matrix recorded and even greater losses. In summary, if we build your home the matrix best option is to buy the same drive, otherwise we risk a large loss of productivity. principle to build a
RAID 0 is very simple, the matrix is about as fast as the slowest drive in it - this should always be remembered.

Last modified on Tuesday, 26 May 2015 16:50
Data Recovery Expert

Viktor S., Ph.D. (Electrical/Computer Engineering), was hired by DataRecoup, the international data recovery corporation, in 2012. Promoted to Engineering Senior Manager in 2010 and then to his current position, as C.I.O. of DataRecoup, in 2014. Responsible for the management of critical, high-priority RAID data recovery cases and the application of his expert, comprehensive knowledge in database data retrieval. He is also responsible for planning and implementing SEO/SEM and other internet-based marketing strategies. Currently, Viktor S., Ph.D., is focusing on the further development and expansion of DataRecoup’s major internet marketing campaign for their already successful proprietary software application “Data Recovery for Windows” (an application which he developed).

4 comments

  • Comment Link Patrick Pay Thursday, 17 November 2016 23:47 posted by Patrick Pay

    I know this if off topic but I’m looking intostarting my own weblog and was wondering what all is required to get setup?I’m assuming having a blog like yours would cost a pretty penny?I’m not very web savvy so I’m not 100 sure. Any tips or advicewould be greatly appreciated. Thanks

  • Comment Link Almeta Coulter Friday, 07 October 2016 13:24 posted by Almeta Coulter

    Just for visiting our website you can get a free copy of our yellow pages scraper and generate yourself a list of leads quickly( limited to 50 free copies ). Here is the download link: http://nichecontactblaster.com/free-yellow-pages-scraper Thanks, http://www.contactformsubmitter.com

  • Comment Link Strople Tuesday, 08 December 2015 08:07 posted by Strople

    liked your post very much

  • Comment Link Strople Tuesday, 08 December 2015 08:07 posted by Strople

    liked your post very much

Leave a comment

Make sure you enter the (*) required information where indicated. HTML code is not allowed.