Seagate's continued expansion into the SSD market hasn't followed the same path as the other storage giants. While many scramble furiously to purchase external companies with existing SSD technology and components, Seagate seems content to leverage existing relationships, and their wealth of internal storage knowledge. Seagate already provides a wide datacenter storage portfolio, and we have evaluated the NAS HDD, 7,200 and 10,000 RPM HDDs, the Enterprise Turbo SSHD, and the 600 PRO 6Gb/s SATA SSDs, which are now joined by the new Seagate 1200 12Gb/s SAS SSD.
The 1200, the third generation of Seagate SAS SSDs, leverages 12GB/s architecture to deliver a high level of performance for a myriad of uses, including virtualization, high-performance computing, data analytics, and tiering.
The design employs the use of Samsung 21nm NAND, which is a rare sight outside of Samsung products. Seagate's use of Samsung NAND is borne of a strategic agreement that provides Seagate the enviable advantage of a guaranteed NAND supply. Seagate augments their access to the fab with OEM and qual-cycle experience, and an extensive global supply chain.
The Seagate 1200 switches from the LAMD controller used in the Pulsar 2 and 600 Pro SSDs, to a custom-designed P6M49400.4 ASIC from Marvell. Seagate's proprietary firmware delivers impressive speeds of 110,000 / 40,000 random 4K read/write IOPS, and sequential performance weighs in at 750 MB/s - 500 MB/s read/write. The 1200 is available in 1.8" or 2.5" form factors with a 7mm z-height, and capacities of 200GB, 400GB, and 800GB.
The 1200 utilizes 512MB of DDR3-1600 cache, along with power capacitors that can flush the contents of the DRAM to the NAND in the event of a power loss. Seagate also focuses on providing robust data protection in the form of SED and FIPS 140-2 models, with support for SANITIZE and cryptographic erase chief, among many security-minded enhancements.
Currently, the only other 12Gb/s SAS SSDs available are offerings from HGST and Toshiba. This leaves us a small comparison pool that is winnowed down to keep the competing SSDs within the same price segment. Many SSD manufacturers are coy about pricing, but Seagate actually has current pricing embedded into their webpage. We previously evaluated the Tohsiba PX02SMF080, and with pricing from between $4-$5 per GB, it is the nearest competitor to the $3-$3.50 per GB price point of the Seagate 1200. Retail prices are generalized and only current at the time of publishing, but are indicative of trends for Tier 2 OEM pricing. As per usual, volume pricing can lead to significant discounts.
Both SSDs have custom Marvell controllers, but the similarities stop there. The Seagate 1200 surprisingly flaunts nearly the same 10 DWPD endurance, with its 21nm Samsung MLC NAND as the Toshiba with its 24nm eMLC. Both drives feature the same five-year warranty, but the Toshiba edges the 1200 in several performance metrics. The Seagate 1200 touts its excellent efficiency with a high IOPS-Per-Watt specification and significantly lower price point.
The stage is set for a showdown between these two 12Gb/s SAS contenders, but first let's take a closer look at the Seagate 1200 family.
Seagate 1200 Architecture
The Seagate 1200 has steady performance specifications for sequential activity with all capacity points, but the 200GB features slightly lower random write performance. The drives are available in 1.8" and 2.5" form factors with a 7mm z-height.
The full-duplex 12Gb/s SAS connection features multi-path ability for failover redundancy to ensure no single point of failure. Advanced SAS error-control enables robust drive-level error correction, and full data path protection. A multi-layered error-recovery technology suite consists of ECC, LDPC (low-Density Parity Check), and Micro-RAID to protect user data. Micro-RAID is likely similar to Marvell's F.R.A.M.E. capability, which provides embedded parity to combat data loss. This combination of technologies provides a read error rate of one LBA in 10E16 bits transferred.
The SSD features an AFR of .44%, and an impressive MTBF of 2 million hours. Capacitors provide a layer of protection from host power loss. User-selectable logical block size (512, 520, 524, 528, 4096, 4160, 4192, or 4224 bytes per logical block) is also supported.
The standard Seagate 1200 features up to 10 drive writes per day (DWPD) of endurance, and the High-Endurance models offer a staggering 25 DWPD. Seagate also provides the option of Lifetime Endurance Management on their Managed Life models. These models allow the user to specify an acceptable write threshold, and throttle the write performance accordingly to keep the SSD within its expected life curve.
SED and FIPS 140-2 compliant SSD models offer Seagate Instant Erase, cryptographic erase for easier drive repurposing and retirement, and the SANITIZE command.
Seagate 1200 Internals
The Seagate 1200 has a very sturdy polished metal case with a bit of heft. The drive is on firmware version 0002.
We note the service port, and a small hole that features a recessed monitoring LED. This monitoring LED flashes in distinct patterns that denote the status of the SSD during operation. Once we remove the top, the thermal pads come into view. The thick pad in the center mates with the PCB under the controller, and effectively sandwiches the controller and assists in wicking away heat. The remainder of the pads cover the NAND and other critical components.
Removal of the other side of the case reveals that the case consists of three parts. Each side of the case holds four screws, and an additional six screws hold the PCB into the center frame. This is a very robust design that is reminiscent of militarized/rugged designs.
The SAS port consists of a long piece of plastic with two fasteners on each end to anchor it to the center frame. In high-vibration environments, the SAS port can become a common source of failures; this sturdy design will likely lead to fewer failures.
The Samsung 21nm MLC NAND is surprisingly resilient with a standard amount of overprovisioning.
The SSD utilizes a custom Marvell controller and 512MB of Micron DDR3-1600 RAM for caching. The banks of power capacitors provide enough capacitance to flush data to the NAND in the event of host power loss. There are open pads for more NAND packages on both sides of the PCB. These additional packages will allow the drive to reach its maximum capacity.
Test System and Methodology
We utilize a new approach to HDD and SSD storage testing for our Enterprise Test Bench, designed specifically to target long-term performance with a high level of granularity. Many testing methods record peak and average measurements during the test period. These average values give a basic understanding of performance, but fall short in providing the clearest view possible of I/O QoS (Quality of Service).
'Average' results do little to indicate performance variability experienced during actual deployment. The degree of variability is especially pertinent, as many applications can hang or lag as they wait for I/O requests to complete. This testing methodology illustrates performance variability, and includes average measurements, during the measurement window.
While under load, all storage solutions deliver variable levels of performance. While this fluctuation is normal, the degree of variability is what separates enterprise storage solutions from typical client-side hardware. Providing ongoing measurements from our workloads with one-second reporting intervals illustrates product differentiation in relation to I/O QoS. Scatter charts give readers a basic understanding of I/O latency distribution without directly observing numerous graphs.
Consistent latency is the goal of every storage solution, and measurements such as Maximum Latency only illuminate the single longest I/O received during testing. This can be misleading, as a single 'outlying I/O' can skew the view of an otherwise superb solution. Standard Deviation measurements consider latency distribution, but do not always effectively illustrate I/O distribution with enough granularity to provide a clear picture of system performance. We utilize high-granularity I/O latency charts to illuminate performance during our test runs.
Our testing regimen follows SNIA principles to ensure consistent, repeatable testing. We measure power consumption during precondition runs. This provides measurements in time-based fashion, with results every second, to illuminate the behavior of power consumption in steady state conditions. Power consumption can cost more over the life of the device than the initial acquisition price of the hardware itself, significantly affecting the TCO of the storage solution. We also present IOPS-to-Watts measurements to highlight efficiency.
Our test pool features SSDs of varying capacity, and it is important to bear this in mind when viewing results. The Toshiba PX02SM series features static performance specifications among its different capacity points. This should provide a fair representation of performance in comparison to the 400GB Seagate 1200. The first page of results will provide the 'key' to understanding and interpreting our new test methodology.
4K Random Read/Write
We precondition the Seagate 1200 for 10,800 seconds, or three hours, receiving reports on several parameters of workload performance every second. We then plot this data to illustrate the drive's descent into steady state.
This chart consists of 18,000 data points. This is a dual-axis chart with the IOPS on the left, and the latency on the right. The red dots signify IOPS during the test, and the grey dots are latency measurements during the test period. We place latency data in a logarithmic scale to bring it into comparison range. The lines through the data scatter are the average during the test. This type of testing presents standard deviation and maximum/minimum I/O in a visual manner.
Note that the IOPS and Latency figures are nearly mirror images of each other. This illustrates that high-granularity testing can give our readers a good feel for latency distribution by viewing IOPS at one-second intervals. This should be kept in mind when viewing our test results below.
This downward slope of performance happens very few times in the lifetime of the device, typically during the first few hours of use, and we present the precondition results only to confirm steady state convergence.
Each QD tested includes 300 data points (five minutes of one second reports) to illustrate performance variability. The line for each QD represents the average speed reported during the five-minute interval.
4K random speed measurements are an important metric when comparing drive performance, as the hardest type of file access for any storage solution to master is small-file random. One of the most sought-after performance specifications, 4K random performance, is a heavily marketed figure.
The Seagate 1200 averages 104,801 IOPS with a 4K random read workload at QD256, while the Toshiba PX02SMF averages an impressive 124,534 IOPS.
Both SSDs provide the best performance to latency ratio at QD128.
Garbage collection routines are more pronounced in heavy write workloads. This leads to more variability in performance. The Seagate 1200 provides an impressive 42,609 IOPS at QD256, with the Toshiba trailing at 28,307 IOPS.
The 1200 provides a much tighter latency distribution than the PX02SMF while under a heavy 4K write workload.
Our write percentage testing illustrates the varying performance of each solution with mixed workloads. The 100% column to the right is a pure write workload of the 4K file size, and 0% represents a pure 4K read workload.
The Seagate features a lower performance envelope from 0-70%, but the extensive variability of the Toshiba muddies the water.
We record the power consumption measurements during our precondition run. We calculate the stated average results during the last five minutes of the test, after the device has settled into steady state.
The 1200 averages 3.94 Watts during the measurement window, and the Toshiba provides a slightly lower average of 4.30 Watts.
IOPS-to-Watts measurements are generated from data recorded during our precondition run, and the stated average is from the last five minutes of the test.
The 1200 bears the fruits of a focus on efficiency with an average of 9,509 IOPS per Watt, while the Toshiba averages 4,835. It is important to note the IOPS per Watts set forth in the specifications of both drives is with read activity. We measure IOPS per Watt with write activity.
8K Random Read/Write
8K random read and write speed is not tested for consumer use, but for enterprise environments this is an important aspect of performance. With several different workloads relying heavily upon 8K performance, we include this as a standard with each evaluation. Many of our Server Emulations below will also test 8K performance with various mixed read/write workloads.
The average 8K random read speed of the Seagate 1200 measures 57,276 IOPS at QD256, while the Toshiba PX02SMF outperforms with an impressive 88,493 IOPS
Both SSDs exhibit a tight latency range during the 8K random read test.
The Seagate 1200 again leads in the heavy write workload test, with an average of 20,403 IOPS at QD256. The Toshiba averages 16,259 IOPS at QD256.
The Toshiba experiences significant latency variability, particularly with the heavier workloads.
The Toshiba PX02SMF again leads with the heavier read workloads, but begins to experience variability quickly once the write workload increases.
Power consumption for the 1200 averages 6.16 Watts during the measurement window, and the Toshiba falls slightly lower with 5.98 Watts.
The Seagate 1200 again leads in the efficiency test with an average of 3,320 IOPS per Watt, while the Toshiba clocks a respectable 2,858 IOPS.
128K Sequential Read/Write
128K sequential speed reflects the maximum sequential throughput of the SSD using a realistic file size encountered in an enterprise scenario. The Toshiba PX02SMF again leads with read activity, mustering an impressive average of 806 MB/s. The Seagate 1200 jumps up to an average of 593 MB/s at QD256.
The Seagate 1200 experiences some variability at QD256, while the Toshiba leverages its impressive sequential read performance with a nice, tight latency distribution.
The 1200 delivers an impressive write speed of 542MB/s at QD256, and the Toshiba PX02SMF delivers 408 MB/s at QD256.
The mixed workload testing exposes significant variability from the Toshiba SSD in mixed sequential workloads.
The Seagate 1200 averages 6.9 Watts during the measurement window, over the 6.09 Watts from the Toshiba PX02SMF.
The Seagate 1200 averages 78 MB/s per Watt, while the Toshiba PX02SMF provides 66 MB/s per Watt.
Database/OLTP and Webserver
This test emulates Database and On-Line Transaction Processing (OLTP) workloads. OLTP is the processing of transactions such as credit cards and high frequency trading in the financial sector. Enterprise SSDs are uniquely well suited for the financial sector with their low latency and high random workload performance. Databases are the bread and butter of many enterprise deployments. These are demanding 8K random workloads with a 66% read and 33% write distribution that can bring even the highest performing solutions down to earth.
The Seagate 1200 averages 38,317 IOPS at QD256, and the Toshiba averages 42,506 IOPS.
The Toshiba experiences significant inconsistency, while the Seagate 1200 operates within a very tight latency envelope.
The Toshiba averages 6.38 Watts, and the Seagate sucks up a miserly 5.28 Watts.
The Seagate 1200 blasts by the Toshiba's 4,103 IOPS with an excellent average of 7,299 IOPS per Watt.
The Webserver profile is a read-only test with a wide range of file sizes. Web servers are responsible for generating content for users to view over the internet, much like the very page you are reading. The speed of the underlying storage system has a massive impact on the speed and responsiveness of the server that is hosting the website, and thus the end-user experience.
The Toshiba PX02SMF leans on its great read performance to master this workload with an average of 43,360 IOPS at QD256. The Seagate 1200 averages a much lower 21,865 IOPS.
The Seagate requires a mere 3.46 Watts for this workload, while the Toshiba draws 5.46 Watts.
The Seagate 1200 loses its only efficiency test of our regimen with an average of 6,378 IOPS, while the Toshiba averages 7,982 IOPS per Watt.
The Emailserver profile is a very demanding 8K test with a 50% read and 50% write distribution. This application is indicative of the performance of the solution in heavy write workloads.
The Seagate delivers a tight average of 30,889 IOPS at QD256. The Toshiba sports an average of 30,748 IOPS, but suffers from significant variability.
The 1200 averages 5.53 Watts in steady state, compared to 6.2 Watts for the Toshiba SSD.
The Seagate closes out our testing with an impressive 5,628 IOPS per Watt, well ahead of the 3,110 IOPS from the Toshiba.
Manufacturers without a guaranteed NAND supply are at a disadvantage, as witnessed with some of the smaller players during the recent NAND shortage. Seagate's unfettered access to Samsung NAND, along with their global distribution channel and OEM relationships, positions them as a potent force in the explosive enterprise SSD market. NAND fabricators have direct access to NAND supply, but often do not enjoy the mature supply chains that Seagate has with their extensive distribution network.
Seagate is willing to use the correct components, combined with their own extensive IP to bolster their SSD product stack. Manufacturing one's own controller is an advantage, but utilizing third party manufacturers is not unlike the approach taken for decades with HDDs. The price of a SSD controller is usually roughly 5-8% of the cost of a SSD, and Seagate has tight control over the design of the controller and proprietary firmware. Seagate's approach should provide more success than those without NAND fabrication capabilities and similar production models.
It is not surprising to find two fab-enabled companies duking it out in the cutting-edge 12Gb/s SAS segment. Both SSDs in the evaluation have strengths, with the Toshiba PX02SMF080 coming out on top in many of the sequential and random read-centric workloads. The Toshiba SSD regularly experienced significant variability in our mixed workloads.
The Seagate 1200 excels in heavy write workloads, and also performed well in our mixed workloads. Average measurements fail to expose the real performance of storage solutions, and our high-granularity testing revealed superb performance consistency from the 1200. Solid consistency performance also bodes well for RAID deployments.
A focus of both drives lies on the efficiency front. The Seagate 1200 nearly swept the board with write efficiency per Watt, falling behind in only one test. We measure write IOPS per Watt, but the Seagate is also spec'd with a higher rating of 29,650 read IOPS per Watt, compared to 13,400 from the Toshiba.
The 1200 takes a multi-layered approach to full data path protection. Capacitors, Micro-RAID, ECC, and LDPC, combine to provide resilience to failure and excellent endurance. Options for enhanced security models, and advanced media management, round out a SSD with an AFR of .44%, an MTBF of two million hours, and a solid five-year warranty.
The Seagate 1200 has the features we have come to expect from enterprise-class SSDs at an aggressive price point that makes it an economical solution for many workloads. The Toshiba PX02SM offers great performance and an impressive series of features as well, albeit at a higher price point, underlining that there is a complex multitude of factors to consider when purchasing an SSD.
Seagate has already gained significant traction with their 600 Pro series, and we expect the 1200 series to find similar success. Samsung is quickly ramping 3D NAND for enterprise applications, and we expect Seagate will provide solutions with this new technology in the future. The Seagate 1200 provides a compelling mix of performance, consistency, efficiency, and endurance, at an outstanding price-point.
PRICING: You can find the Seagate 1200 SSD and the Toshiba PX02SMF080 for sale below. The prices listed are valid at the time of writing, but can change at any time. Click the link to see the very latest pricing for the best deal.
United States: The Seagate 1200 SSD (200GB) retails for $760.00 at Amazon, and the Toshiba PX02SMF080 (800GB) retails for $3,215.00 at Amazon.