October 1st, 2007 by eugene
Faceoff at One Terabyte: Seagate's Barracuda ES.2 and WD's Caviar GreenPower
Leveraging a unique five-platter design, Hitachi Global Storage managed to bring the formidable Deskstar 7K1000 to the market well before competing designs. For several months now, Hitachi's beast has combined the best capacity and performance one could get on the SATA interface. Now, however, competitors Seagate and Western Digital have commenced shipment of their first terabyte units... and each manufacturer's take is a bit different from that of Hitachi's.
The Barracuda ES.2 is the enterprise-grade version of Seagate's consumer-oriented Barracuda 7200.11 and the successor to the firm's 750-gigabyte Barracuda ES 750 (retrospectively the "ES.1"). With the luxury of a later introduction, the ES.2 manages its terabyte capacity utilizing just four 250-gigabyte platters as contrasted to the Deskstar 7K1000's five-disc design. Seagate specs the ES.2's seek time at 8.5 milliseconds.
With the ES.2, Seagate joins Hitachi as one of the first manufacturers to offer a 32-megabyte buffer on its drive as an option across the entire family's capacity. The Deskstar 7K1000's enviable performance gains suggest that the time may be ripe for the industry as a whole to move to this new level. ES.2 units featuring a more standard 16-megabyte cache are also available.
The Barracuda ES.2 is also the first enterprise-class SATA design from Seagate that will also be available equipped with a Serial Attached SCSI (SAS) interface. Slowly but surely SAS is displacing the traditional 68-/80-pin LVD interface with its vastly simplified architecture and its convenient interoperability with the mass-market SATA connection.
Seagate equips the ES.2 with a vibration compensation system typically associated with enterprise drives to mitigate performance problems that arise when many drives are packed into the same chassis and rack. The manufacturer also boasts an MTBF and error recovery rate that rival the best of the SCSI/SAS world (1.2 million hours and 1x10E-15 respectively). Seagate backs the unit with a 5-year warranty.
Also with the ES.2, Seagate commences with a newer focus on environmental concerns. Reducing power draw nets a double-edged savings through both reduced power consumption itself and reduction in dissipated heat and associated cooling costs. Seagate claims the improvements, collectively offered under its "PowerTrim" label, net a 20% reduction in draw when contrasted with the drive's predecessor... and 55% in watts-per-gigabyte when the ES.2's larger capacity is taken into account.
WD's offering in the terabyte sweeps is the WD10EACS, or the Caviar "GreenPower" (GP). Like Seagate, WD takes advantage of additional development time to offer a one-terabyte capacity incorporating four 250-gigabyte platters rather than the five 200-gigabyte discs found in the Hitachi Deskstar 7K1000. The firm claims an 8.9 millisecond seek time and equips the GP with a standard 16-megabyte buffer. A three-year warranty backs the drive.
The focus, as one can guess from the drive's name, is squarely on environmental concerns. WD, more so than Seagate, banks on power consumption issues becoming more and more a significant part of overall drive ownership and operation concerns as time passes. Through an aggressive deployment of a variety of features, the GP line boasts of power dissipation of up to 40% less than baseline standards, a figure that WD claims will save up to $10 per year per drive when considered under 24x7 operation.
The manufacturer is careful in not directly citing spindle speed, instead nominally positioning the Caviar GP as a "7200 RPM-class" drive. Under its "IntelliPower" moniker, WD claims a "A fine-tuned balance of spin speed, transfer rate and cache size designed to deliver both significant power savings and solid performance." Some folks have misinterpreted some admittedly vague specs on WD's website. Under "Rotational Speed," the manufacturer cites "IntelliPower (5400 to 7200 RPM)." This does not mean the drive dynamically changes its spindle speed during operation... indeed, such a feature would entail considerable mechanical engineering and would in many ways defeat the point -- rapidly accelerating and decelerating the spindle's speed would increase rather than decrease net power draw. Rather, the IntelliPower term indicates that the GP family as a whole does not have a set spindle speed (nor a set buffer size, for that matter). Different capacity points may feature differing spin speeds and buffer sizes. For those that must know, WD admits "sub-6000 RPM operation" for the 1-TB Caviar GP (more on this on the following page).
WD argues that, like processor speeds and the associated clock speed specs that dominated CPU marketing for years, disc drive focus should move away from individual low-level specs/claims and instead focus on end results. For our part, we agree. A couple years back, we moved the StorageReview Leaderboard away from listing drives by "spindle speed class" to the current "intended use" paradigm.
The Caviar also incorporates the just-in-time (IntelliSeek) paradigm originally introduced in WD's notebook drives and "IntelliPark," a system where the drive's read/write heads are powered down and moved away from the platter assembly when not in use to reduce aerodynamic drag.
Seagate's 1-TB Barracuda ES.2 (with 32-megabyte buffer option) and WD's 1-TB Caviar GP WD10EACS will be compared against the following drives for the following reasons:
|Hitachi Deskstar 7K1000 (1000 GB)||Industry's First Terabyte Drive|
|Seagate Barracuda ES (750 GB)||Predecessor to the Barracuda ES.2|
|WD Caviar SE16 WD7500AAKS (750 GB)||Predecessor to the WD Caviar GP|
|WD Raptor WD1500ADFD (150 GB)||Performance-oriented 10K RPM SATA reference drive|
For diagnostic purposes only, StorageReview measures the following low-level parameters:
Average Read Access Time- An average of 25,000 random read accesses of a single sector each conducted through IPEAK SPT's AnalyzeDisk suite. The high sample size permits a much more accurate reading than most typical benchmarks deliver and provides an excellent figure with which one may contrast the claimed access time (claimed seek time + the drive spindle speed's average rotational latency) provided by manufacturers.
Average Write Access Time- An average of 25,000 random write accesses of a single sector each conducted through IPEAK SPT's AnalyzeDisk suite. The high sample size permits a much more accurate reading than most typical benchmarks deliver. Due to differences in read and write head technology, seeks involving writes generally take more time than read accesses.
WB99 Disk/Read Transfer Rate - Begin- The sequential transfer rate attained by the outermost zones in the hard disk. The figure typically represents the highest sustained transfer rate a drive delivers.
WB99 Disk/Read Transfer Rate - End- The sequential transfer rate attained by the innermost zones in the hard disk. The figure typically represents the lowest sustained transfer rate a drive delivers.
For more information, please click here.
The Barracuda ES.2 weighs in with a 12.7 millisecond (ms) seek time, about a half-millisecond above the drive's predecessor. Accounting for the drive's 7200 RPM spindle speed (4.2 ms latency) leaves the unit with a measured access time of 8.5 ms... bravo Seagate, right on the money. The ES.2 joins the ES 750 now as the second consecutive Seagate offering that meets the firm's claimed specs... a welcome turnaround from a company that used to publish overly optimistic claims that often were missed by its ATA/SATA drives.
We have to take a different approach when assessing the Caviar GP's seek time since WD does not explicitly give users the drive's spindle speed. The GP turns in a measured access time of 15.0 ms, a score that lags the 7200-RPM WD7500AAKS by a significant margin. The WD7500AAKS's measured seek time when accounting for 4.2 ms of 7200 RPM latency is 9.5 ms (missing the firm's claim by over half a millisecond). Assuming the GP also shares such a seek time, that leaves us with 15 ms [measured access time] minus 9.5 ms [assumed seek time] which equals 5.5 ms, almost exactly the rotational latency associated with a 5400 RPM spindle speed.
Seagate's latest is the first SATA drive to bust through the 100 MB/sec plateau through achieving 104 MB/sec of transfer in its outermost zones. Rates decay down to a more pedestrian 53.3 MB/sec.
The Caviar GP's outer-zone score clocks in at 79.8 MB/sec and as a result lags the older, less dense WD7500AAKS by 21%. Assuming similar sector-per-track zone configurations, a 7200 RPM drive would boast a 33% advantage over a 5400 RPM unit. The difference between the GP and the WD7500AAKS is less than that, likely of course due to a density advantage on the GP's part. Nevertheless, this second low-level diagnostic again suggests a 5400 RPM spindle speed.
It is important to remember that seek time and transfer rate measurements are mostly diagnostic in nature and not really measurements of "performance" per se. Assessing these two specs is quite similar to running a processor "benchmark" that confirms "yes, this processor really runs at 2.4 GHz and really does feature a 400 MHz FSB." Many additional factors combine to yield aggregate high-level hard disk performance above and beyond these two easily measured yet largely irrelevant metrics. In the end, drives, like all other PC components, should be evaluated via application-level performance. Over the next few pages, this is exactly what we will do. Read on!
StorageReview uses the following tests to assess non-server use:
StorageReview.com Office DriveMark 2006- A capture of VeriTest's Business Winstone 2004 suite. Applications include Microsoft's Office XP (Word, Excel, Access, Outlook, and Project), Internet Explorer 6.0, Symantec Antivirus 2002 and Winzip 9.0 executed in a lightly-multitasked manner.
StorageReview.com High-End DriveMark 2006- A capture of VeriTest's Multimedia Content Creation Winstone 2004 suite. Applications include Adobe Photoshop v7.01, Adobe Premiere v6.5, Macromedia Director MX v9.0, Macromedia Dreamweaver MX v6.1, Microsoft Windows Media Encoder 9.0, Newtek Lightwave 3D 7.5b, and Steinberg Wavelab 4.0f run in a lightly-multitasked manner.
For more information, please click here.
WD's Caviar GP turns in a StorageReview Office DriveMark of 794 I/Os per second (IOps), lagging its predecessor by about 10%.
The Seagate Barracuda ES.2's showing of 588 IOps builds very incrementally on the score achieved by the ES 750, an almost negligible 2% increase.
In our higher-level, content-creation-oriented High-End DriveMark, both newcomers stumble a bit when compared to previous models. The WD10EACS lags the older WD7500AAKS by almost 13%. Perhaps more surprisingly, however, the Barracuda ES.2 regresses when contrasted with the ES 750, trailing its predecessor by 15%.
Three decidedly different entertainment titles cover gaming performance in StorageReview's test suite.
FarCry, a first-person shooter, remains infamous for its lengthy map loads when switching levels.
The Sims 2, though often referred to as a "people simulator," is in its heart a strategy game and spends considerable time accessing the disk when loading houses and lots.
Finally, World of Warcraft represents the testbed's role-playing entry; it issues disk accesses when switching continents/dungeons as well as when loading new textures into RAM on the fly.
For more information, please click here.
The Barracuda ES.2 builds upon the ES 750 with a modest but measurable 6% gain in our replay of FarCry's disc access. The Caviar GP, on the other hand, trails the WD7500AAKS by 11%.
The Sims 2 places the GP about 15% behind the WD7500AAKS. Seagate's latest also slips slightly, lagging its forerunner by about 2%.
It is World of Warcraft that places the biggest gap between WD's two drives. The GP lags the WD7500AAKS by 18% here. Meanwhile, the Barracuda ES.2 regains some form and manages a 13% lead over the ES 750.
Unlike single-user machines (whether a desktop or workstation), servers undergo highly random, non-localized access. StorageReview simulates these multi-user loads using IOMeter. The IOMeter File Server pattern balances a majority of reads and minority of writes spanning requests of varying sizes.
IOMeter also facilitates user-configurable load levels by maintaining queue levels (outstanding I/Os) of a specified depth. Our tests start with the File Server pattern with a depth of 1 and double continuously until depth reaches 128 outstanding I/Os.
Drives with any sort of command queuing abilities will always be tested with such features enabled. Unlike single-user patterns, multi-user loads always benefit when requests are reordered for more efficient retrieval.
For more information click here.
The Barracuda ES.2 delivers substantial improvement over the ES 750 when it comes to our multi-user test. While both drives commence with nearly identical results under a linear load, the ES.2 scales significantly better under concurrency to top the chart when it comes to 7200 RPM units. The ES.2's improvements approach a commendable 40% over the relatively modest ES 750.
The Caviar GP unsurprisingly lags its performance-oriented predecessor by margins of approximately 10%. The gap closes as queue depths hit extremely heavy levels, however.
Noise and Power Measurements
Idle Noise- The sound pressure emitted from a drive measured at a distance of 3 millimeters. The close-field measurement allows for increased resolution between drive sound pressures and eliminates interactions from outside environmental noise. Note that while the measurement is an A-weighted decibel score that weighs frequencies in proportion to human ear sensitivity, a low score does not necessarily predict whether or not a drive will exhibit a high-pitch whine that some may find intrusive. Conversely, a high score does not necessarily indicate that the drive exhibits an intrusive noise profile.
Operating Power Dissipation- The power consumed by a drive, measured both while idle and when performing fully random seeks. In the relatively closed environment of a computer case, power dissipation correlates highly with drive temperature. The greater a drive's power draw, the more significant its effect on the chassis' internal temperature.
Startup (Peak) Power Dissipation- The maximum power dissipated by a drive upon initial spin-up. This figure is relevant when a system features a large number of drives. Though most controllers feature logic that can stagger the spin-up of individual drives, peak power dissipation may nonetheless be of concern in very large arrays or in cases where a staggered start is not feasible. Generally speaking, drives hit peak power draw at different times on the 5V and 12V rails. The 12V peak usually occurs in the midst of initial spin-up. The 5V rail, however, usually hits maximum upon actuator initialization.
For more information, please click here.
WD's latest turns in an idle sound pressure measurement of just 35.9 dB/A, a score that ties it with the Samsung SpinPoint T166 (a smaller 500-gigabyte not included in this review's charts... full scores, as always, are available from the SR Performance Database) as the quietest drive we have ever measured. Subjectively speaking, the GP's noise floor is utterly amazing. It is easily the quietest ever at idle, eerily silent even outside of a case. Seeks are also whisper quiet, significantly lower than all other modern 7200 RPM drives.
The Barracuda ES.2's measured pressure of 42.0 dB/A places it slightly above the ES 750's recorded score. Subjectively, however, the ES.2's idle noise floor is lower than that we perceive from the ES 750, though not as quiet as the Caviar GP. Seeks remain on par with those of the ES 750, noticeable but unobtrusive when grinding away.
At 3.8 watts, the Caviar GP delivers the lowest measured power dissipation by far of any drive that has met Testbed4, a reduction of 54% when contrasted with the WD7500AAKS. Under a full load, the power savings remain with the GP weighing in well under 8 watts... a figure lower than that of any other 3.5" drive at idle.
When compared to its predecessor, the Barracuda ES.2 shaves one watt off when idle and manages 11.4 watts when fully seeking. The latter represents a 17% improvement, just shy of meeting Seagate's claim of a 20% difference.
The ES.2 trims about 7 watts off of the ES 750's peak current draw when spinning up from a cold start, a decent improvement that places Seagate's newest in the upper echelon when it comes to power-up draw. WD's drive manages a more modest one watt improvement over the WD7500AAKS.
According to filtered and analyzed data collected from participating StorageReview.com readers, a predecessor of the Seagate Barracuda ES.2, the Seagate Barracuda 7200.9 , is more reliable than of the other drives in the survey that meet a certain minimum floor of participation.
According to filtered and analyzed data collected from participating StorageReview.com readers, the Western Digital Caviar GP is more reliable than of the other drives in the survey that meet a certain minimum floor of participation.
According to filtered and analyzed data collected from participating StorageReview.com readers, a predecessor of the Western Digital Caviar GP, the Western Digital Caviar WD4000KD , is more reliable than of the other drives in the survey that meet a certain minimum floor of participation.
Note that the percentages in bold above may change as more information continues to be collected and analyzed. For more information, to input your experience with these and/or other drives, and to view comprehensive results, please visit the SR Drive Reliability Survey.
What this means for the Barracuda 7200.11 (a virtually identical drive marketed to consumers rather than the enterprise) is unclear, but with the ES.2, Seagate has succeeded in its goal of offering a server-oriented terabyte drive that offers top-notch performance while trimming power draw to lower levels.
WD Caviar GP WD10EACS
The Caviar GP stands as an iconoclast when compared with the rest of the SATA field. 5400 RPM drives went out of vogue several years ago yet the advantages of a lower spindle speed have possibly become more important than ever before in this age of environmental awareness. As a result, the firm's desire to shy away from publishing traditional specs is understandable. Citing "IntelliPower" rather than "lower RPM" makes sense from a marketing standpoint given the drive's goal: power consumption in a class of its own without sacrificing performance. Has it succeeded?
WD's latest trails the older 750 GB Caviar WD7500AAKS by margins of 11%-17% when it comes to StorageReview's single-user performance tests. Given that WD has traditionally been a firm that pushed the envelope when it comes to the speed that enthusiasts crave, the GP falling short of a speed-oriented design is unsurprising and perhaps inevitable. Of course, what is considered "fast" and "slow" can be relative. After all, the Caviar GP trounces the 7200 RPM, 32-megabyte buffer Barracuda ES.2 in single-user tests.
The GP does easily draw the least amount of power of any 3.5" drive we've ever tested... less than 1/2 of any other unit when idle. Even under a full seek, the GP enjoys an advantage of over 30% over the typical 7200 RPM SATA drive. In what almost seems like an ancillary effect, the GP also happens to be the quietest drive we have ever measured and heard here at SR. Its miniscule power dissipation also translates into far less need for cooling which yields yet more acoustic benefit. Especially when taken in the context of arrays featuring many drives that run 24x7, lower cost of operation also becomes significant.
So, those seeking the ultimate in speed and those inclined to judge solely by low-level specifications probably will not find the Caviar GP on their short list. However, those who have always found that concentrating on power, heat, and noise issues always meant sacrificing on high capacities and/or competitive performance may just find their soul mate in WD's GreenPower drive.