Seagate’s Barracuda ES hard drive

Manufacturer Seagate
Model Barracuda ES
Price (street)
Availability Now
SCSI HAS LONG RULED the enterprise world, but Serial ATA is slowly creeping into corporate server rooms. Serial ATA simply can’t be beaten when it comes to storage density, and when you’re paying for not only the drives, but also the rack they sit in, density can be an even more important metric than performance.

Today, the highest capacity SCSI drives top out at 300GB—just 60% of the capacity of today’s enterprise-class Serial ATA drives, which are available up to 500GB. Seagate has just raised the bar even higher, introducing a new Barracuda ES hard drive with a whopping 750GB of storage. To put that into perspective, consider the storage capacity of a four-drive 1U rack server running RAID 5. With 300GB SCSI drives, you won’t even break one terabyte. 500GB Serial ATA drives could give you 1.5TB of redundant capacity, while an array of 750GB Barracuda ES drives would offer a cool 2.25TB, all in the same physical space.

The Barracuda ES owes its freakish capacity to the perpendicular recording technology it shares with Seagate’s desktop-oriented Barracuda 7200.10. Physically, the drives are nearly identical. However, the Barracuda ES packs firmware optimizations that promise better performance under more demanding loads, an attribute whose appeal extends beyond the enterprise world and into the enthusiast’s realm.

How does the Barracuda ES compare with other enterprise-class Serial ATA hard drives? Is it really any faster than the Barracuda 7200.10? Read on to find out.

What makes an ES?
The Barracuda ES is based on the same drive platters and mechanics as the 7200.10, but it’s built to withstand cramped rackmount enclosures where drives can be packed tighter than steerage class on a budget airline. All hard drives vibrate during normal spinning and seek operations, and those vibrations can disrupt the operation of a drive in close proximity by shaking the drive head off its intended path. The disrupted drive must then wait for its head to move back into position before resuming normal operation, resulting in a performance penalty.

To combat vibration-induced performance degradation, the Barracuda ES is outfitted with sensors that detect rotational vibration and adjust the drive head accordingly. These sensors allow the ES to tolerate a rotational vibration of 12.5 rad/second2 with a rotational profile of 20-800Hz. The Barracuda 7200.10, on the other hand, can only tolerate 5.5 rad/second2 between 10 and 300Hz.

  Barracuda 7200.10 Barracuda ES
Maximum external transfer rate 300MB/s 300MB/s
Sustained transfer rate NA 78MB/s (750GB)
72MB/s (500, 400, 250GB)
Average read seek time NA 8.5ms
Average write seek time NA 9.5ms
Average rotational latency 4.16 ms 4.16ms
Spindle speed 7,200RPM 7,200 RPM
Available capacities 200, 250, 300, 320, 400, 500, 750GB 250, 400, 500, 750GB
Cache size 8MB (200GB)
8/16MB (250GB)
16MB (320-750GB)
8/16 MB (250GB)
16MB (400-750GB)
Platter size 188 GB (750GB) 188 GB (750GB)
Rotational vibration 5.5 rad/sec2 12.5 rad/sec2
Rotational vibration profile 10-300Hz 20-800Hz
Idle acoustics 2.7 bels 2.7 bels
Seek acoustics 3.0 bels 3.0 bels
Idle power consumption 9.3W 9.3W
Read/write power consumption 12.6W 13.0W
Native Command Queuing Yes Yes
Warranty length Five years Five years

Apart from the ES’s rotational vibration sensors, the drive is physically identical to the Barracuda 7200.10. It uses the same 188GB platters with perpendicular recording and is available with up to 16MB of cache. That cache is programmed a little differently on the ES, of course, but it’s the same chip.

In addition to different cache segmentation, the ES’s firmware offers a handful of RAID-specific features that you won’t find in the Barracuda 7200.10. Among them is Error Recovery Control, which limits the drive’s error recovery time to 12 seconds. This prevents prolonged error recovery attempts from causing drives to be prematurely dropped from RAID arrays. Similar functionality is available on Western Digital’s enterprise-class Serial ATA drives, although WD recommends disabling the feature for single-drive operation. Error Recovery Control apparently doesn’t hinder the Barracuda ES in single-drive systems.

Error recovery is nice, but avoiding errors is even better. To help reduce the number of errors the ES encounters in thermally challenging enterprise environments, the Barracuda ES monitors temperatures and shuffles I/O requests in a manner that allows the drive to cool if it gets too hot. The ES also includes features to simplify multi-drive firmware updates and speed RAID initialization.

Although both the Barracuda ES and 7200.10 are available with a 300MB/s Serial ATA interface, the ES also comes in fibre channel flavors. Those flavors are limited to 400 and 500GB capacities with only 8MB of cache, though.

Hard drive manufacturers seem loath to publish performance specifications for their desktop hard drives, so it’s hard to compare the ES’s seek times and sustained transfer rate specs to those of the 7200.10. We’ll see how real world seek times and transfer rates shake out when we move to our performance benchmarks.

Enterprise-class Serial ATA drives typically offer longer warranties than their desktop counterparts—often five years for enterprise models versus just three years for desktop. Seagate’s a little different, though. It offers a five-year warranty for all its internal hard drive products, including desktop and enterprise drives, so the ES doesn’t have an advantage over the 7200.10 in this category. Five-year hard drive warranties are pretty standard in the enterprise world, so Seagate certainly isn’t skimping. The company’s enthusiasm for longer desktop drive warranties has just blunted what could have been another selling point for the ES.

 
Test notes
We’ll be comparing the performance of the Barracuda ES with that of a slew of competitors, including some of the latest and greatest Serial ATA drives from Hitachi, Maxtor, Seagate, and Western Digital. These drives differ when it comes to external transfer rates, spindle speeds, cache sizes, platter densities, NCQ support, and capacity, all of which can have an impact on performance. Keep in mind the following differences as we move through our benchmarks:

  Max external transfer rate Spindle speed Cache size Platter size Capacity Native Command Queuing?
Barracuda 7200.7 NCQ 150MB/s 7,200RPM 8MB 80GB 160GB Yes
Barracuda 7200.8 150MB/s 7,200RPM 8MB 133GB 400GB Yes
Barracuda 7200.9 (160GB) 300MB/s 7,200RPM 8MB 160GB 160GB Yes
Barracuda 7200.9 (500GB) 300MB/s 7,200RPM 16MB 125GB 500GB Yes
Barracuda 7200.10 300MB/s 7,200RPM 16MB 188GB 750GB Yes
Barracuda ES 300MB/s 7,200RPM 16MB 188GB 750GB Yes
Caviar SE16 300MB/s 7,200RPM 16MB 83GB 250GB No
Caviar SE16 (500GB) 300MB/s 7,200RPM 16MB 125GB 500GB Yes
Caviar RE2 150MB/s 7,200RPM 16MB 100GB 400GB Yes
Caviar RE2 (500GB) 300MB/s 7,200RPM 16MB 125GB 500GB Yes
Deskstar 7K500 300MB/s 7,200RPM 16MB 100GB 500GB Yes
DiamondMax 10 150MB/s 7,200RPM 16MB 100GB 300GB Yes
Raptor WD740GD 150MB/s 10,000RPM 8MB 37GB 74GB No*
Raptor X 150MB/s 10,000RPM 16MB 75GB 150GB Yes
Raptor WD1500ADFD 150MB/s 10,000RPM 16MB 75GB 150GB Yes

Note that the 250GB Caviar SE16 and the Raptor WD740GD lack support for Native Command Queuing. The WD740GD does support a form of command queuing known as Tagged Command Queuing (TCQ), but host controller and chipset support for TCQ is pretty thin. Our Intel 955X-based test platform doesn’t support TCQ.

We have test results from older and newer versions of Western Digital’s Caviar SE16 and RE2. To avoid confusion, we’ll be referring to the newer drives as the Caviar RE2 (500GB) and Caviar SE16 (500GB), while the old drives will appear as the Caviar RE2 and Caviar SE16.

Since Seagate makes versions of the 7200.7 both with and without NCQ support, the 7200.7 in our tests appears as the “Barracuda 7200.7 NCQ” to clarify that it’s the NCQ version of the drive. The Caviar RE2, Deskstar T7K250, DiamondMax 10, 7200.8, 7200.9, 7200.10, ES, Raptor X, and Raptor WD1500ADFD aren’t explicitly labeled as NCQ drives because they’re not available without NCQ support.

Finally, we should note that our WD1500ADFD has a slightly newer firmware revision than the Raptor X sample we’ve had since February. The drives still share identical internals, but firmware optimizations could give our newer Raptor an edge over the X in some tests.

Our testing methods
All tests were run three times, and their results were averaged, using the following test system.

Processor Pentium 4 Extreme Edition 3.4GHz
System bus 800MHz (200MHz quad-pumped)
Motherboard Asus P5WD2 Premium
Bios revision 0422
North bridge Intel 955X MCH
South bridge Intel ICH7R
Chipset drivers Chipset 7.2.1.1003
AHCI/RAID 5.1.0.1022
Memory size 1GB (2 DIMMs)
Memory type Micron DDR2 SDRAM at 533MHz
CAS latency (CL) 3
RAS to CAS delay (tRCD) 3
RAS precharge (tRP) 3
Cycle time (tRAS) 8
Audio codec ALC882D
Graphics Radeon X700 Pro 256MB with CATALYST 5.7 drivers
Hard drives Hitachi 7K500 500GB SATA
Maxtor DiamondMax 10 300GB SATA
Seagate Barracuda 7200.7 NCQ 160GB SATA
Seagate Barracuda 7200.8 400GB SATA
Seagate Barracuda 7200.9 160GB SATA
Seagate Barracuda 7200.9 500GB SATA
Seagate Barracuda 7200.10 750GB SATA
Western Digital Caviar SE16 250GB SATA
Western Digital Caviar RE2 400GB SATA
Western Digital Raptor WD740GD 74GB SATA
Western Digital Raptor X 150GB SATA
Western Digital Raptor WD1500ADFD 150GB SATA
Western Digital Caviar RE2 500GB SATA
Western Digital Caviar SE16 500GB SATA
Seagate Barracuda ES 750GB SATA
OS Windows XP Professional
OS updates Service Pack 2

Our test system was powered by OCZ PowerStream power supply units. The PowerStream was one of our Editor’s Choice winners in our last PSU round-up.

We used the following versions of our test applications:

The test systems’ Windows desktop was set at 1280×1024 in 32-bit color at an 85Hz screen refresh rate. Vertical refresh sync (vsync) was disabled for all tests.

All the tests and methods we employed are publicly available and reproducible. If you have questions about our methods, hit our forums to talk with us about them.

 

WorldBench overall performance
WorldBench uses scripting to step through a series of tasks in common Windows applications. It then produces an overall score. WorldBench also spits out individual results for its component application tests, allowing us to compare performance in each. We’ll look at the overall score, and then we’ll show individual application results alongside the results from some of our own application tests.

Seagate’s perpendicular-powered Barracudas are the fastest 7,200-RPM drives in WorldBench, but interestingly, the ES is one step slower than the 7200.10.

Multimedia editing and encoding

MusicMatch Jukebox

Windows Media Encoder

Adobe Premiere

VideoWave Movie Creator

Scores are close through WorldBench’s multimedia editing and encoding tests. The Barracuda 7200.10’s usually a little faster than the ES, though.

 
Image processing

Adobe Photoshop

ACDSee PowerPack

ACDSee separates the drives a little and relegates the ES to the middle of the pack. There, it’s still a few seconds slower than the 7200.10.

Multitasking and office applications

Microsoft Office

Mozilla

Mozilla and Windows Media Encoder

WorldBench’s office and multitasking tests don’t give the drives much room to stretch their legs.

Other applications

Winzip

Nero

WinZip and Nero do, however. The ES is close to the front of the pack in both, with only Western Digital’s latest 10K-RPM Raptors and the Barracuda 7200.10 proving faster.

 
Boot and load times
To test system boot and game level load times, we busted out our trusty stopwatch.

Boot times on the ES are very fast, but the drive’s performance in our level load tests isn’t as impressive.

 
File Copy Test
File Copy Test is a pseudo-real-world benchmark that times how long it takes to create, read, and copy files in various test patterns. File copying is tested twice: once with the source and target on the same partition, and once with the target on a separate partition. Scores are presented in MB/s.

The Barracuda ES stumbles a little in FC-Test’s file creation tests, losing ground to a couple of Raptors, and more importantly, Western Digital’s latest 7,200-RPM Caviars. Note that the ES lags behind the 7200.10 by a few MB/s with each test pattern.

Seagate bounces back with a stronger showing in FC-Test’s read tests, outrunning the Caviars and even the Raptors in a couple of test patterns. The performance gap between the ES and 7200.10 is much smaller here.

The copy tests’ combination of read and write operations suits the ES well enough to keep it ahead of the competition with most test patterns. However, the Caviars have a slim edge with the Windows and Programs test patterns, which feature a greater number of smaller files. Interestingly, those are the same test patterns where the ES trails the 7200.10 by the greatest margins.

 
iPEAK multitasking
We’ve developed a series of disk-intensive multitasking tests to highlight the impact of command queuing on hard drive performance. You can get the low-down on these iPEAK-based tests here. The mean service time of each drive is reported in milliseconds, with lower values representing better performance.

The ‘cuda ES starts our iPEAK tests with a win, but soon falls victim to the same Jekyll-and-Hyde performance profile that plagues all of Seagate’s Barracudas in these tests. With multitasking loads that feature a file copy operation as the secondary task, the ES performs very well. However, when that secondary task switches to a VirtualDub import, the ES is ejected rather abruptly to the back of the field with the rest of its Barracuda brethren.

Note that the ES has generally trailed the 7200.10 thus far, but it has a consistent edge in this first round of iPEAK tests.

 
iPEAK multitasking – con’t

The same patterns persist through our second round of iPEAK tests. The ES does well with one set of secondary loads, but poorly with another. Throughout, however, it maintains a slight edge over the 7200.10.

 
IOMeter – Transaction rate
IOMeter presents a good test case for command queuing, so the NCQ-less Western Digital Caviar SE16 and Raptor WD740GD should have a slight disadvantage here under higher loads.

The sheer number of results make these graphs a little hard to read, but I wanted to include them all to illustrate just how many drives perform better than the Barracuda ES in IOMeter. Obviously, the 10K-RPM Raptors are way out ahead, but Western Digital’s enterprise-oriented Caviar RE2s also have a big edge over the ES, especially under increasing loads.

What’s most disappointing, perhaps, is the fact that the Barracuda ES isn’t dramatically faster than the 7200.10. Here’s how the two compare without all the other drives clouding the picture.

The ES isn’t all that much faster in any test pattern, and doesn’t hit its stride until the number of outstanding I/O requests starts to add up.

 
IOMeter – Response time

Moving to IOMeter response times, the Barracuda ES again finds itself well behind the Raptors and Western Digital’s latest Caviars. Heck, even the older Barracuda 7200.7 NCQ is quicker than the ES here.

Again, I’ve whipped up a second set of graphs to highlight how the 7200.10’s performance compares with that of the ES.

And again, the enterprise-oriented ES doesn’t offer much of an improvement over its desktop counterpart. Response times are consistently lower, but not by a significant margin.

 
IOMeter – CPU utilization

To its credit, the Barracuda ES’s CPU utilization is very low in IOMeter. Then again, that’s true for all the drives we tested.

 
HD Tach
We tested HD Tach with the benchmark’s full variable zone size setting.

The ES can’t quite catch the 7200.10, but its sustained transfer rates easily beat the rest of the 7,200-RPM field. Sustained transfer rates are usually dictated by a combination of spindle speed and platter size, and with higher capacity platters than any other 7,200-RPM drive, it’s easy to see why the ES does so well.

Seagate drives have always fared well in HD Tach’s burst speed test, and the ES is no exception. The drive doesn’t quite max out the interface’s theoretical 300MB/s capacity, but it comes closer than drives from other manufacturers.

Enterprise-class hard drives are usually known for quicker access times, but not the Barracuda ES. In HD Tach, its 14.1 millisecond access time is slower than the rest of the field, and a half-millisecond off that of the 7200.10.

CPU utilization scores are well within HD Tach’s +/- 2% margin for error in this test.

 
Noise levels
Noise levels were measured with an Extech 407727 Digital Sound Level meter 1″ from the side of the drives at idle and under an HD Tach seek load. Drives were run with the PCB facing up.

The Barracuda ES is about as quiet as the 7200.10 at idle, but a decibel louder under a seek load. That makes it one of the loudest drives of the bunch, but noise tends not to be a problem in server rooms already polluted by the constant hum of industrial air conditioning.

Power consumption
For our power consumption tests, we measured the voltage drop across a 0.1-ohm resistor placed in line with the 5V and 12V lines connected to each drive. Through the magic of Ohm’s Law, we were able to calculate the power draw from each voltage rail and add them together for the total power draw of the drive.

Seagate has managed to improve the Barracuda ES’s power consumption over the 7200.10, but there are still a number of drives that consume less juice. If we consider power consumption per gigabyte, however, the Barracuda ES fares better than any other drive.

 
Conclusions
With online vendors selling the Barracuda ES for around $465, the drive’s price tag certainly lives up to its enterprise aspirations. A Barracuda 7200.10 with the same 750GB capacity can be had for closer to $340, and for most folks—performance-hungry enthusiasts included—that’s the drive to have. For desktop applications, the ES just doesn’t offer enough over the 7200.10 to justify the price premium.

The question, of course, is whether that price premium is justified for the enterprise environments that Seagate is targeting with the drive. From a performance perspective, the answer is no. The Barracuda ES is too slow under multitasking and multi-user loads to compete with Western Digital’s Raptors or even WD’s latest enterprise-oriented Caviar RE2s—drives that feature similar RAID-specific error recovery and vibration compensation capabilities.

Of course, neither the Raptors nor the Caviars are available with 750GB of storage capacity. Freakish capacity is the ‘cuda’s real appeal, especially when you’re trying to squeeze as many gigabytes as possible into limited enclosure or rack space. 500GB is the best the competition can do on that front, and with 50% more capacity in the same 3.5″ form factor, a 750GB Barracuda ES certainly has some attractiveness. It may not be the fastest drive in its class, but it’s the biggest, and in some enterprise environments that will matter more. 

Comments closed
    • Dissonance
    • 13 years ago

    Added additional details on some of the ES’s RAID-specific features. Our overall conclusions remain unchanged, though.

    • albundy
    • 13 years ago

    soooo…wut yur sayin is dat the cuda has the same qualities of a cheetah? I doubt any corporation will be willing to bet their data and existence on that. Although, this setup is perfect for a 10 employee company.

    • flip-mode
    • 13 years ago

    aw crap

    • leor
    • 13 years ago

    i wonder whos’ gonna be the first to launch a tera drive?

    • Cannyone
    • 13 years ago

    I allways thought “Enterprise” class drives were intended for multi-drive setups (some form of RAID array). Now I understand the problems with performing your review that way, as has also been noted by Dposcorp. Still I feel you did an excellent job. Of course some of that opinion is biased by my personal preference for Seagate drives. 😉 (In particular I’m personally trying to acquire more 7200.10s immediately.)

    Oh and in reply to #1 – RAID 5 uses what’s called “rotating parity”. This means that the capacity of one drive is distributed in a staggered stripe across all of the drives, which allows this type of array to rebuild itself if one of the drives should fail. At least that’s how I understand it! (I’ve been wrong in the past…)

    • Dposcorp
    • 13 years ago

    Nice review, Diss.

    I figured that the performance would be close to the same as the desktop model.

    Since we are taking enterprise here, I am a bit surprised we did not see some RAID array testing with a quality controller. (RAID 5 would be nice to see)

    Of course, I do know that you would then need 4 drives of each make, which is expensive to get and time consuming to setup.

    Still, it would be interesting to see how these drives perform in the type of environment they appear to be marketed for, against some of the other drives you tested.

    Also, I have a issue with this sentence.

    q[<“The company's enthusiasm for longer desktop drive warranties has just blunted what could have been another selling point for the ES.”<]q Maybe it is just me, but I rather see Seagate get kudos for bringing their desktop drives up to 5 year warranties, rather then the above statement, which gives the appearance of being a knock against Seagate. I’ll admit to being a huge Seagate fan, and have purchased 4 of their drives just last month.

      • IntelMole
      • 13 years ago

      Since not everyone offers massive desktop warranties, in a head to head between the two drives this is pretty accurate.

      As for RAID testing – surely you’re testing the RAID controller as much as the RAID drive then?

        • Dposcorp
        • 13 years ago

        q[

          • indeego
          • 13 years ago

          Controllers can act differently based on the drive/RAID combination. I’ve seen many a buggy Controller BIOS on cheapo desktop MB’sg{<.<}g

            • Dposcorp
            • 13 years ago

            Tthat is true, as i remember reading maxtors had problems with some Nv controllers.

            But that will come up during testing, and those drives/arrays can be removed from that test if need be.

            If a lot of them have problems, then I would blame the controller.

            All I am saying is, a nice 3 or 4 drive array test would be nice to see, from 2-3 different brands of drives.

            • Buub
            • 13 years ago

            So don’t use cheapo desktop RAID controllers for an enterprise drive comparison. Stick an LSI or 3ware controller in there and use that, as a good server builder would.

            • indeego
            • 13 years ago

            Why? I don’t buy desktops so that I can put anything mission critical on there. But the idea of raid1 on a desktop is sexy to me because it means a much greater recovery speed if it works like it should. By far the biggest hassle I deal with as an admin is recovery after a failed drive for desktops or freaking as a raid 5 or 1 array is rebuilding for servers. Because the chance is there during the rebuild that it could failg{<.<}g

            • BobbinThreadbare
            • 13 years ago

            I think he was refering to benchmark purposes only.

      • eitje
      • 13 years ago

      i agree – if the article is going to make reference to how the real value of this drive is the added capacity for a RAID 5 in a 1U server, i’d appareciate knowing how the performance compares, as well. 🙁

    • The Swamp
    • 13 years ago

    1, The 4th drive is a hot spare. We have a SCSI RAID 5 setup at our office that is configured this way.

    • flip-mode
    • 13 years ago

    Two errors on front page:

    g[

      • flip-mode
      • 13 years ago

      Hm, still there. Am I incorrect?

        • Damage
        • 13 years ago

        Fixed ’em earlier. Sorry. Forgot to say anything.

      • flip-mode
      • 13 years ago

      Waoh. Not there anymore 🙂

    • stmok
    • 13 years ago

    Just a minor criticism on the article. On page 11, could you possibly re-scale those charts, such that we can see a better picture of the results?

    For example: The first graph on that page, would it be possible to re-scale such that the max is 1%

    Granted, it doesn’t matter all that much, but it provides a consistant look with the rest of the article. (Instead of a clump of lines at the bottom of each chart).

    Other than that minor issue, the article is good.

    I wonder if its possible to come up with a future test to really gauge the reliability of HDDs. 🙂

    • Namarrgon
    • 13 years ago

    /[

      • Dissonance
      • 13 years ago

      /[

        • eitje
        • 13 years ago

        oooooo!!! RAID knowledge burn!

Pin It on Pinterest

Share This