How do we test?

Testenvironment

We recommend that readers who aren't interested in test procedures, jump over the following page and head directly to the test results.

Model	Samsung Spinpoint F3 HD103SJ	1'000 GByte
Motherboard	Gigabyte X38-DQ6
Chipset	Intel X38	1066 MHz
CPU	Intel Xeon 3060	2.4 GHz
Memory	Corsair Dominator 9136 DDR2	2 GByte
Graphics card	Gigabyte GeForce 7200
Storage (system)	Maxtor	160 GByte
Operating systems	Ubuntu 8.10 Hardy	Kernel 2.6.27-11-server
Filesystem	XFS

We think everybody reading this article can imagine the following scenario: You just bought a hard drive, which according the specs sheet should transfer 120 MByte/s reading and writing. In the reviews you read about astonishing 110 MByte/s but after you put the drive into your system it feels much slower. The whole story gets even worse when you start a benchmark which does random read/write of 4 KByte blocks. There you only get two to three MBytes/s.

Because of this we don't want to publish screenshots of standard programs like HD-Tach, HD-Tune, ... we want our tests to be

• reproducible,
• accurate
• meaningful and
• varied

We test with activated caches and NCQ (Native Command Queueing) because they're also activated in daily use condition. The amount of data when the drive is tested is at least twice as much as the amount of memory in the testsystem.

We noticed that the measuring error is constantly within ±2%. Therefore we mention it only here.

Additionally we evaluate the S.M.A.R.T. data to assess if there are already errors.

The following table gives you a brief overview to which points we turn our centre of attention.

Test	Observations
Sequential Read/Write Tests	Are the values within the specifications? Which influence has the block size? Which influence has the filesystems block size?
Random Read/Write Tests	How severe is the influence on the theoretically possible (sequential) datarate? Which influence has the block size? Which influence has the block size on the filesystem?

Linux, XFS, why?

There are different reasons why we choose an operating system based on a Linux Kernel instead of a fresh Windows Vista/XP installation with all Service Packs.

• The filesystem XFS offers you flexibility you don't get from NTFS.
• The testprogram iozone runs natively under linux and offers a wide variety of possibilities.
• The test partly aims to server applications.
• Iozone gives you data on a statistical basis (error, deviation, etc.)

The filesystem is going to be built as follows:

mkfs.xfs -f /dev/sdb

We mounted with the following options:

mount -o rw,noatime,logbufs=8 /dev/sdb /mnt/hdd

The throughput is being measured using the following parameters:

iozone -Rb test_xk.out -i0 -i1 -i2 -+n -r xk -s4g -t2 

The operations per second are being measured using the following parameters:

iozone -Rb test_xk.out -i0 -i1 -i2 -+n -r xk -s4g -t2 -O

Why do we test different block sizes?

It is important to reproduce daily usage scenarios. Certain parameters need to be variable during the test to make a statement about the product. In our test the those parameters are the different block sizes. It defines the size in KBytes which is written/read on the drive during a transaction.

With this method one can test the reading and writing of either small or big files. In a normal personal computer environment you usually don't find files smaller than 16 KByte because the smallest amount of data which is written matches the block size. Even if your operating system wants to write/read an only 0.5 KByte small file a whole 16 KByte block is read. Furthermore the standard blocksize in Windows XP is 16 KByte.

In bigger RAID arrays the hard disk cache is usually disabled and the RAID-Controller takes over the job of caching. Exactly in such setups hard drives need to be very fast when reading or writing small amounts of data. Sequential throughput isn't interesting in this case.

Page 1 - Introduction	Page 5 - Random read/write KByte/s
Page 2 - Preview	Page 6 - Sequential write/read ops
Page 3 - How do we test?	Page 7 - Random write/read ops
Page 4 - Sequential write/read KByte/s	Page 8 - Conclusion

Discuss this article in the forum