iXBT Labs - Computer Hardware in Detail

Platform

Video

Multimedia

Mobile

Other

Seagate Barracuda 180 hard disk



By Kirill Kochetkov

In many fields of the computer business the situation is defined with a principle "one has to run very fast to remain at the same place". The leader's position is really very shaky, and if he stops for a bit of rest, the competitors will outperform him right away.

Seagate company decided to strengthen its position in the high-efficient HDD market with a release of the biggest in size hard disc Barracuda 180. The 180 GBytes disc can store twice as much data comparing with the latest developments.

The situation is similar to that of NVidia company on the video accelerator market. Today, the power of the latest NVIDIA GeForce3 chip can't still find its user since programs (i.e. games) do not use all its possibilities, and the modern processors are not able to prepare complex scenes for it so fast. In case of the Barracuda 180, the majority of modern SCSI solutions use 18 and 36 GBytes discs and 5-10 times leap is not very interesting for them, especially considering that they usually use RAID system at least with three discs to ensure high reliability and speed. Undoubtedly, there is some difference - the Barracuda is intended for serious, corporate applications (I doubt a user would take a $2000 disc for his home PC), and NVIDIA is working mainly for home gamers.

In this review we will examine what other possibilities the new disc gives, and whether it is a revolutionary jump or an obvious step in the course of hard disc development.

Technical parameters:

Barracuda 180

  • size: 180 GBytes
  • interface: Ultra 160 SCSI, Fibre Channel
  • spindle rotation speed: 7200 RPM
  • latency: 4.17 ms
  • average read/write seek time: 7.4/8.2 ms
  • average read/write track to track seek time: 0.8/1.1 ms
  • internal formatted data rate: 26-47 Mbytes/s
  • buffer size: 4 (optionally 16)Mb
  • discs/heads: 12/24
  • noise (idle): 3.7 bels
  • power consumption (idle): 10.3 W
  • dimensions (HxWxD): 40.64 x 101.6 x 146.05 mm
  • weight: 1.043 kg

The disc is produced in 6 variants: there are 3 interface versions - Ultra160 ("LW"), Ultra160 SCA ("LC") and Fibre Channel ("FC"), and models with different buffer size - 4 or 16 MBytes (index "V") (the latter is positioned for stream video applications). We had a model for Ultra160 interface with 4 MBytes buffer (ST1181677LW). By the way, the names of these Seagate models consist of 7 figures what gets more difficult to remember :).

Barracuda 180 front

Barracuda 180 rear

So, let's start with capacity. The 180 GBytes size allows for more than 2.5Tb disc space on one SCSI controller. It is 2.5 times more than earlier (on 73 GBytes discs). Plus, usage of one disc for achieving 180 GBytes disc space is an effective solution. Of course, it is only hour and a half of unpacked video of high quality, but it is still much more than it was before. Now up the price. The disc wins from the previous generation of the Barracuda 18XL in cost per MB; the manufacturer recommended price is $2150 (and in online stores it can be taken at $1700!), and the 18 GBytes ST318436LW disc costs $280. It means that the Barracuda 180 is 1.5 cheaper considering the cost per MB. Power consumption: comparing with the Barracuda 18XL, we have it almost 6 times less for one Barracuda 180 disc as compared with 10 Barracuda 18XL ones. As for noise, the situation is also better - 3.7 bels as compared with 3.2-3.4 for modern 7200 RPM discs.

It seems that by the 7th generation of Barracuda discs the technology of their production is so perfect that it was just modesty that didn't allow the Seagate engineers to install not 12 but, say, 17 discs in a 1.6" case and to write the figure 256 in the disc name.

On the one hand, combination of an increase in size, reduction of the price per MB, of power consumption and noise is a very big step forward, and on the other hand, a 20% makeweight doesn't seem to be so hard to reach, considering that Seagate was producing discs with 10 platters before. So, let's draw a conclusion: Seagate has managed to combine all modern technological achievements in the Barracuda 180 disc, what has helped to create a product according to the status of this company. Note that modern IDE discs contains 20 GBytes platters, that is why we might see Barracuda 240 very soon. By the way, there is another interesting version of their development - the Barracuda 180 printed circuit board is more similar to the Cheetah X15 board (the picture is here) than to its predecessors - Barracuda 18XL and 50.

Let's see what place the Baracuda 180 takes among HDDs with SCSI interface:

First up the official data:

Quantum
Atlas V
Seagate
Barracuda 18XL
IBM
Ultrastar
36LP
Seagate
Barracuda 180
capacity (GBytes)
9.1
18.3
36.7
9.2
18.4
9.1
18.3
36.9
180
RPM
7200
7200
7200
7200
average seek time (ms)
6.3
5.9
6.8
7.4
average track to track seek time (ms)
0.8
0.6
0.6
0.8
latency (ms)
4.17
4.17
4.17
4.17
Internal Data Rate /
Sustained Throughput (Mb/sec)
17-29
17.2-29.6
19.5-31.9
26-47
buffer size (MBytes)
4
2
4
4/16
interface
Ultra160
Ultra160
Ultra160
Ultra160
height (inches)
1.00
1.0
1.0
1.6
noise (Bels, Idle)
3.2
3.3
3.5
3.4
-
3.7
power consumption (W, Idle)
7.5
8.7
9.9
6.5
6.8
7.2
7.9
8.9
10.3

 

Quantum
Atlas 10k II
Seagate
Cheetah 18XL,
36LP,
73
IBM
Ultrastar
36LZX
IBM
Ultrastar
72ZX
Seagate
Cheetah X15
capacity (GBytes)
9.2
18.4
36.7
73.4
9.2
18.4
36.7
73.4
9.1
18.3
36.7
73.4
9.2
18.4
RPM
10000
10000
10000
10000
15000
average seek time (ms)
4.7
4.7
4.7
5.2
5.2
5.2
5.2
5.6
4.9
5.3
3.9
average track to track seek time (ms)
0.6
0.6
0.5
0.6
0.5
latency (ms)
3.0
2.99
2.99
2.99
2.0
Internal Data Rate /
Sustained Throughput (MBytes/sec)
24-40
26.6-40.5
21.5-36.1
22.1-37.4
38.0-47.4
buffer size (MBytes)
8
4
4
16
4
interface
Ultra160
Ultra160
Ultra160+
Ultra160+
Ultra160
height (inches)
1.0
1.0
1.0
1.64
1.0
1.0
1.0
1.6
1.0
1.6
1.0
noise (Bels, Idle)
3.7
3.7
3.8
4.2
3.8
3.8
3.9
4.2
-
-
3.9
power consumption (W, Idle)
9.7
9.7
10.0
14.3
8.0
9.0
11.0
15.7
8.5
9.7
12.9
16.9
11.0

 

As you can see, the Barracuda 180 easily competes even with 10,000 RPM discs in some parameters. But since the speed parameters are measured by companies' own methods, we shouldn't be guided only by them.

Tests

The test system consists of the ASUS CUV4X-DLS (with LSI Ultra160 SCSI controller), the Intel Pentium III 800EB processor, 256 MBytes PC133 memory. The programs were run under the Windows 2000 Pro SP1.

From the ZDLabs Winbench 99 1.2 set we have taken only Disk Transfer Rate and Disk Access Time tests. These results depend mainly on the disc (not too much on a controller) that is why we have taken some figures from the previous testing of SCSI discs for comparison (note than the speed is measured in thousands of bytes per second):

Atlas
V
Barracuda
18XL
Barracuda
180
Atlas
10k2
Cheetah
18XL
Ultrastar
36LZX
Cheetah
X15
size
18 GBytes
9 GBytes
180 GBytes
18 GBytes
9 GBytes
36 GBytes
18 GBytes
RPM
7200
10000
15000
Transfer Beginning (TB/s)
30000
27000
42400
42000
36000
36000
42000
Transfer End (TB/s)
17000
16000
24000
28000
24000
22000
32000
Disk Access Time (ms)
12.1
12
12.2
7.28
9.04
8.67
6.80
graph of reading

The Barracuda 180 has quite a strange graph - the transition from one zone into another is followed by considerable losses in speed. It may strongly affect the quality of working with discs, for example, with video streams (more than 20 MBytes/s may be required there for one stream of unpacked video). The Cheetah X15 is in a similar situation (but with less losses). Perhaps, it is connected with the fact that we testes a pre-production sample, and production ones will lack for it (other tests of a linear reading speed have shown analog results).

It is interesting that in case of the Quantum, the linear reading speed is the same in the both tables, and the Seagate has the ratings overstated. However, it is curious that the Barracuda 180 and Cheetah X15 have the rated max speed equal to 47 MBytes/s, and the tests showed it to be 40 MBytes/s (the same(!) in the both cases). Of course, Seagate calls this parameter "Transfer rate Internal Formatted (Mbytes/sec)", but an end-user has nothing to do with an internal data rate, he is mostly interested in an external speed, the one that can be measured and compared.

The Disk Transfer Rate and Disk Access Time tests may be enough to define the overall performance of the disc. Of course, the speed is also influenced by the buffer size, working algorithms, the internal electronic system, an operating system, a file system and others but the data of these two tests are the most important. And according to these data, the Barracuda 180 is much faster than its 7200 RPM brothers, and in the Disk Transfer Rate it competes against even the 10000 RPM discs.

But today integral tests based on real applications are also popular. The problem is that is is difficult to get a strong competitor for this disc. Besides, I doubt anybody would run programs included, for example, in the ZDLabs Business/High-End Disk Winmark (business applications, image processing, Internet, multimedia).

That is why the second part of the test is some kind of an artificial character. It was conducted with usage of the Intel IOMeter test (we have written about it here). For comparison we have taken a leader of the previous tests in 10,000 RPM class - the Quantum Atlas 10k2. By the way, for SCSI discs with 7200 RPM the maximum volume before was only 50 GBytes (Barracuda 50). Apart from such well known patterns as Database, File Server, Workstation we used patterns for random/successive reading/writing of data. Let me remind you parameters of all patterns once more:

pattern's name size % of size % reads % random
Database
8192
100
67
100
File Server
512
10
80
100
1024
5
80
100
2048
5
80
100
4096
60
80
100
8192
2
80
100
16384
4
80
100
32768
4
80
100
65536
10
80
100
Workstation
8192
100
80
80
Data Streaming Read
8192
100
100
0
Data Streaming Write
8192
100
0
0
Data Random Read
8192
100
100
100
Data Random Write
8192
100
0
100

The first part is Database, File Server, Workstation:

IOMeter group 1

The random data access was depicted on the second picture:

IOMeter group 2

And the last part is successive data reading and writing:

IOMeter group 3

In the first two parts the Barracuda 180 lags behind at the expense of a bigger access time. Of course, the discs have different spindle speeds, but the Barracuda 180 leads in the size*speed/price parameter.

The third picture is more interesting - despite of a higher read speed and a bigger buffer size, the Atlas 10k2 has considerably lagged behind the Barracuda 180. But the figures are not very accurate - each pattern has processed around 20 GBytes of data. It means that the 18 GBytes Atlas 10k2 has passed the test completely and has received an average score, whereas the Barracuda 180 had time to process only a tenth part. I think that exactly this affected the final result. Unfortunately, we had the sample of the Barracuda 180 for not a long time, and, therefore, we failed to conduct the test in corpore.

As for noise, the disc is at the level of modern 7200 RPM discs. And as for heat dissipation, I can say that it warms up stronger that its 7200 RPM brothers but weaker than 10,000 RPM discs. That is why I recommend to use additional cooling.

Conclusion

With the Barracuda 180 disc Seagate company has proved its leading position not only in production of new solutions but also in usage of the well-known technologies. The Barracuda 180 disc will undoubtedly take its place in those systems where a greater size and the high reliability of data storage are required. In addition, it should be noted that Seagate gives a 5 year warranty.

Write a comment below. No registration needed!


Article navigation:



blog comments powered by Disqus

  Most Popular Reviews More    RSS  

AMD Phenom II X4 955, Phenom II X4 960T, Phenom II X6 1075T, and Intel Pentium G2120, Core i3-3220, Core i5-3330 Processors

Comparing old, cheap solutions from AMD with new, budget offerings from Intel.
February 1, 2013 · Processor Roundups

Inno3D GeForce GTX 670 iChill, Inno3D GeForce GTX 660 Ti Graphics Cards

A couple of mid-range adapters with original cooling systems.
January 30, 2013 · Video cards: NVIDIA GPUs

Creative Sound Blaster X-Fi Surround 5.1

An external X-Fi solution in tests.
September 9, 2008 · Sound Cards

AMD FX-8350 Processor

The first worthwhile Piledriver CPU.
September 11, 2012 · Processors: AMD

Consumed Power, Energy Consumption: Ivy Bridge vs. Sandy Bridge

Trying out the new method.
September 18, 2012 · Processors: Intel
  Latest Reviews More    RSS  

i3DSpeed, September 2013

Retested all graphics cards with the new drivers.
Oct 18, 2013 · 3Digests

i3DSpeed, August 2013

Added new benchmarks: BioShock Infinite and Metro: Last Light.
Sep 06, 2013 · 3Digests

i3DSpeed, July 2013

Added the test results of NVIDIA GeForce GTX 760 and AMD Radeon HD 7730.
Aug 05, 2013 · 3Digests

Gainward GeForce GTX 650 Ti BOOST 2GB Golden Sample Graphics Card

An excellent hybrid of GeForce GTX 650 Ti and GeForce GTX 660.
Jun 24, 2013 · Video cards: NVIDIA GPUs

i3DSpeed, May 2013

Added the test results of NVIDIA GeForce GTX 770/780.
Jun 03, 2013 · 3Digests
  Latest News More    RSS  

Platform  ·  Video  ·  Multimedia  ·  Mobile  ·  Other  ||  About us & Privacy policy  ·  Twitter  ·  Facebook


Copyright © Byrds Research & Publishing, Ltd., 1997–2011. All rights reserved.