Toshiba's launch of the PX02SM series of eSSDs also brings about the unveiling of 12Gb/s SAS connectivity for storage devices. Through a lucky set of circumstances, and an undisclosed 12Gb/s adapter on loan from an R&D department, we happen to have an 800GB PX02SMF080 on hand and the necessary equipment to give us a peek into the coming world of 12Gb/s SSD's. The dual-port 12Gb/s SSD features sequential read/write speeds of 900/400 MB/s respectively, and random read/write IOPS of 120,000/30,000.
Last week Toshiba announced full availability of the PX02SMx eSSD (enterprise SSD's) for European markets. Understandably, this caught us by surprise as there are currently no 12Gb/s RAID adapters or HBA's on the market. Our early samples of the PX02SMx series have been languishing in the enterprise storage lab for nearly a month, while we awaited the key component for testing a 12Gb/s SSD: a 12Gb/s adapter. All 12Gb/s adapters from the normal vendors are under NDA or currently only sampling to OEMs. Luckily, we had already set plans in motion to gain access to the necessary equipment for testing.
The launch of 12Gb/s SAS brings a doubling of bandwidth that on the surface seems to be the most compelling feature. PCIe 3.0 and SAS 12Gb/s seem to be the perfect match, but in reality, the insatiable speed provided by 12Gb/s SAS already outstrips the available bandwidth of PCIe 3.0.
12Gb/s also brings other enhancements and changes to the SAS storage fabric that will allow for an expansion of the SAS ecosystem. SAS has long been an interface just begging to be let out of the box, and with the inclusion of new connectivity features, SAS is finally able to expand beyond its current limitations. We will cover some of these improvements on the following pages.
The major players with NAND foundries are all now rapidly expanding their presence in the enterprise realm with a variety of tiered solutions that appeal to different environments. The Toshiba PX02SMF080 series of eSSDs is the leading edge of a full frontal assault from Toshiba into the enterprise SSD battleground.
The PX02SM aims for Tier 0 applications with a dual-port 12Gb/s SAS interface and is Toshiba's first eSSD to feature 24nm eMLC (enterprise Multi Level Cell) NAND. The entry and mid-level PX02AM series also features 24nm eMLC NAND, but utilizes a SATA connection. The low-cost PX03AN series rounds out the offerings by leveraging standard 19nm MLC NAND and a SATA connection for entry-level and read-centric applications. All three SSD's support cryptographic-erase functionality in addition to AES 256-bit self-encryption.
The SED versions of the PX02SM series, the PX02SMQ/U, also offer Trusted Computing Group (TCG) enterprise protocol self-encryption. These features are important in today's computing environment where, according to Symantec, the average loss of an enterprise data breach is $5.5 million.
The PX02SM flagship product we are evaluating today features a Toshiba/Marvell controller in conjunction with 24nm eMLC and comes in capacities of 200GB, 400GB, 800GB and 1,600GB with a slim 7mm Z-height. The PX02SM series also features longevity that scales with capacity, with up to 29.2PB of write endurance for the 1,600GB model, which translates to roughly 10 DWPD (Drive Writes Per Day). This endurance promise is backed by a five year warranty.
The PX02SM also utilizes Toshiba's proprietary Quadruple Swing-By Code (QSBC) to protect from read errors, an integral component of a layered ECC approach for improved error correction capabilities. The addition of power loss protection defends data from power loss events.
The Px02SMF080 is designed to support mission critical enterprise applications that demand performance, endurance and frugal power consumption. Toshiba has extensive history and experience with flash memory; they actually invented NAND flash over 25 years ago in 1987. Leveraging these decades of experience brings forth a serious contender to the enterprise SSD market.
Toshiba PX02SM Specifications
Toshiba PX02SM Specifications Specifications
The Toshiba PX02SMx series features 24nm eMLC NAND in conjunction with a Toshiba/Marvell TC58NC9036GTC controller with custom firmware. The SSD provides 120,000/30,000 random read/write IOPS and 900/400MB/s of sequential read/write speed.
There is also the inclusion of 1GB of DRAM caching to manage LBA mapping on the 800GB, 256MB for the 200GB and 512MB for the 400GB model. The PX02SM series is compatible with 6Gb/s and 12Gb/s SAS dual port connections. The dual port SAS connection provides multipath and failover redundancy.
The PX02SMx series SSDs, and the 800GB PX02SMF080 we are evaluating today, support SANITIZE Cryptographic-Erase functionality and 256-bit AES encryption. Unlike a regular disk erase using lengthy over-write operations, the Cryptographic-Erase function simply regenerates the SED drive's encryption key, effectively invalidating all previously stored user data. This allows SED storage devices to be quickly and securely sanitized before re-allocation, redeployment or retirement.
The PX02SMQ/U series also supports the TCG Enterprise Security Subsystem Class (SSC) protocols to provide additional security features.
The PX02SM comes in capacities of 200GB, 400GB, 800GB and 1,600GB with a 7mm Z-height (the 1.6TB version has a 15mm Z-height). The endurance of the SSD scales with capacity and endurance of 3.7PB, 7.3PB, 14.6PB and 29.2PB is warrantied for up to five years for the various capacity points. This equates to roughly 10DWPD of endurance.
Power loss protection is built-in with a bank of tantalum capacitors to flush data in transit down to the NAND in the event of a power failure. The energy efficiency is 13,400 IOPS per Watt, though this is measured as read IOPS per Watt. A maximum draw of 9 Watts is listed as the peak power consumption.
One missing element from the chart is the UBER rating of 1 in 10E17.
SAS 12Gb/s Architecture
The emergence of mega-datacenters and high-density deployments has required more connectivity features from SAS, while the simultaneous explosion of flash into the datacenter has required massive performance improvements at a pace unseen in the recent history of storage technology.
The regular update cadence of the SAS interface has provided a massive performance jump of 6-10X in the last three years. This rapid upgrade schedule and wide deployments will help SAS survive the challenge from new emerging specifications. While these new specifications have a number of appealing enhancements, they are still in their infancy with no working silicon in general availability.
One of the keys to success for the SAS interface has been existing infrastructure and backwards compatibility. The lack of backward compatibility, and infrastructure changes that could be required in some cases, can hinder widespread adoption of any new interface. Many refer to SAS as the 'Ethernet of storage'. Like Ethernet, SAS has been challenged by many different protocols over the years, but fast upgrade cycles and support for existing infrastructure have helped it weather these storms.
12Gb/s SAS is now entering general availability with HBA's and RAID controllers, and the new storage products to connect to them, being released over the next several months. The evolution to 12Gb/s SAS will continue for several years. The first 24Gb/s Plugfest is slated to begin in 2016, followed by end-user availability in 2017. This gives 12Gb/s SAS roughly four years before it is superseded by a more robust SAS protocol.
Aside from the bandwidth expansion, which allows 12Gb/s to fully saturate the PCIe 3.0 bus at 8,000MB/s, the upgraded architecture also brings new connectors that provide more expansive connection capabilities and enhanced density.
The new Mini-SAS HD (High Density) SFF-8643/8644 connections provide both an active and passive scheme for 6Gb/s and 12Gb/s SAS. These new connectors also sport an electrically improved design with less cross-talk and a better signal to noise ratio.
Adaptec recently switched to the HD connectors with their line of 6Gb/s products and other RAID and HBA vendors are expected to make the switch to Mini-SAS HD with 12Gb/s products. The smaller connector allows the connection of more devices to the smaller HHHL (Half Height, Half Length) controllers that are becoming commonplace in today's dense server deployments.
The expanded speed of 12Gb/s SAS will also raise the current cap of 24GBps throughput of a normal 4 lane SAS cable up 48GBps. This will enhance density by allowing the connection of more devices per expander before the connection between the HBA/RAID controllers becomes the bottleneck.
The SAS Connectivity Roadmap also points to a massive upgrade in cabling distances, with 20 meters for active copper and up to 100 meters with optical connections. This will provide a means for SAS to support box-to-box, server-to-storage, and rack-to-rack connections. In conjunction with SAS switches and expanders, active copper and optical can create an entirely new ecosystem within the datacenter. These new capabilities will enable expanded topologies for both custom builders and modular applications.
Improving the connectivity and ease of operation is a tall order, but the new Mini-SAS HD connectors deliver a spate of enhanced functionality. SAS Connectivity Management supports connection discovery and cable management by detecting the presence (or absence) of passive, active, and optical connections. Along with rapid fault isolation and the minimization of configuration errors, this can simplify large-scale deployments significantly.
The only initial inhibitor for Mini-SAS HD will be the significantly higher price of cabling for the next few months. As we hit general availability and the economies of scale factor in, there should be significant easing of the pricing structure. Overall, the widespread adoption of these new connections and features with the 12Gb/s generation of storage devices will simplify and expand the capabilities of SAS; allowing for greater distance, scalability, usability and serviceability for large-scale deployments.
Toshiba PX02SMF080 Internals
The Toshiba PX02SMF080 comes in a 2.5" form factor with a 7mm Z-height. The 1.6TB versions come with a 15mm Z-height. The case is constructed of a lightweight metal alloy and features the relevant branding. The top of the case also holds a recessed series of indentations that function as heat channels. The rear of the case has a metal recessed area, and we noted that there is a square hole, and several round ones, on the bottom of the SSD that exist to allow heat to flow from the interior of the case.
The thermal characteristics of the heat channels become clear when we open the case of the SSD. There are small thermal pads that mate with the NAND and DRAM chips to wick heat away from these components. The small pads are present on both sides of the case, four small thermal pads mate with the top of the controller on one side, and another four connect to the PCB on the other side of the controller. The thermal pads are much smaller than the components, as evidenced by the oily residue they leave on the NAND packages.
We tend to observe the presence of very large thick thermal pads that cover entire banks of NAND in many enterprise SSDs, but in this type of 'open' case, this could impede the internal airflow. The previous generation 6Gb/s SAS SLC MK4001GRZB featured these same type of holes in the case to provide internal airflow.
Interestingly enough, the protrusions on the internal heat channels are indented on the other side of the case. In other SSD designs these type of channels are extended into the interior while the exterior of the case remains flat. This may be due to aerodynamics that allow for more efficient heat transfer on the exterior of the SSD with recessed channels.
The large controller dominates the top of the PCB. There are 8 x 64GB NAND packages (8K page) on both the top and bottom of the PCB with 8 x 64Gb die per package. This gives us 1024GB of raw NAND, though only 745GB is user addressable.
There are two DDR3 DRAM packages near the controller on both sides of the PCB. This provides the eSSD 512MB of DRAM for a fast area of memory to hold LBA mapping tables. The 200GB drive has 256MB, and the 800GB drive carries 1GB of DRAM.
The banks of tantalum capacitors line the top of the PCB on the edge are also on the bottom of the PCB. There are no extra 'pads' for connection of more capacitors on this 800GB eSSD. With many SSDs, the low capacity versions will have fewer capacitors, but leave open pads for manufacturing to add more capacitors to scale with the increased number of components with higher capacity versions. This 800GB SSD is half the capacity of the largest 1.6TB version, and with the larger version having double the thickness at 15mm, it would be safe to assume that the largest PX02SM features a dual PCB design. There is no visible connection hardware for another PCB, so the layout will likely be significantly different for the 1.6TB model.
The etched Toshiba marking has an almost holographic appearance to it, making it hard to photograph. The controller also features a Marvell branding, and a part number that doesn't match with any Marvell controllers we have experience with. The 6GB/s MK4001GRZB predecessor sported a controller with a 88SS9032-BLN2 part number. The TC58NC9036GTC on the 12Gb/s SAS SSD seems to employ a different numbering convention, but the jump from a '9032' to a '9036' version makes sense. Toshiba is holding details of this controller, including core and channel count, close to the chest. The SSD is employing Toshiba proprietary firmware, but other details are scarce.
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 the 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 use histograms to illuminate the latency of every single I/O issued during our test runs.
Our testing regimen follows SNIA principles to ensure consistent, repeatable testing. We attain steady state through a process that brings the device within a performance level that does not range more than 20% during the measurement window. Forcing the device to perform a read-write-modify procedure for new I/O triggers all garbage collection and housekeeping algorithms, highlighting the real performance of the solution.
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. This significantly affects the TCO of the storage solution. We also present IOPS-to-Watts measurements to highlight the efficiency of the storage solution.
Our test pool features SSDs of varying capacity, so it is important to bear this in mind when viewing results. With the first 12Gb/s SAS SSD in hand, there is no other 12Gb/s SSD that we can compare to that isn't under NDA. We chose the 6Gb/s SMART Optimus in a dual port configuration on the 12Gb/s controller as the other SAS contender. The 12Gb/s Toshiba PX02SMF080 was also tested in a dual port configuration on the same SAS 12Gb/s adapter.
The first page of results will provide the 'key' to understanding and interpreting our new test methodology.
4K Random Read/Write
We precondition the Toshiba PX02SMF080 for 18,000 seconds, or five hours, receiving reports on several parameters of workload performance every second. We then plot this data to illustrate the drives' descent into steady state.
This chart consists of 36,000 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 the point that high-granularity testing can give our readers a good feel for the latency distribution by viewing IOPS at one-second intervals. This should be in mind when viewing our test results below.
We also provide histograms for further latency granularity. 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 for every parameter tested includes 300 data points (five minutes of one second reports) to illustrate the degree of 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 Toshiba PX02SMF080 produces a blistering average of 124,224 IOPS at QD256, followed by the SMART Optimus with an average of 100,290 IOPS at QD256 in dual port mode. The Optimus has a tighter read pattern with less variability, but with the Toshiba blasting away with 24,000 more IOPS, the variability isn't a concern.
Garbage collection routines are more pronounced in heavy write workloads, leading to more performance variability. The Toshiba PX02SMF080 does not quite reach the beastly random write performance of the SMART Optimus, which averages 44,326 IOPS at QD256. The Toshiba 12Gb/s chugs along with a respectable average of 27,165 IOPS at QD256, but significant performance variability muddies the performance picture. The outlying I/O's that reach down to 15,000 IOPS lower the average speed. Some of this variability might be introduced with the use of dual port functionality. However, the Optimus is tested with the same parameters and does not experience as much variability.
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 PX02SMF080 leads the pack with the pure random read workload, but falls into range of the Optimus as we mix in more write activity. The Optimus actually overtakes the PX02SMF080 in the 90% and 100% write categories by leveraging its higher random write performance.
The Toshiba PX02SMF080 provides 2,557,840 I/O's (31.49%) in the 10-20ms range, 1,766,687 I/O's (21.75%) at 6-8ms, 1,544,610 I/O's (19.02%) at 8-10ms, and 1,153,118 I/O's (14.20%) at 2-4ms.
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 Toshiba sports significantly lower power consumption than the Optimus, with an average of 5.63 Watts in comparison to the Optimus' measurement of an average of 7.48 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 Toshiba PX02SMF080 averages 4,843 write IOPS per Watt in comparison to the Optimus with 5,906 write IOPS per Watt.
8K Random Read/Write
8K random read and write speed is a metric that 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 Toshiba PX02SMF080 comes in at 88,120 IOPS at QD256. The Optimus manages an average of 74,240 IOPS at QD256, but peaks at 80,000 IOPS at QD64.
The average 8K random write speed of the Toshiba is 15,715 IOPS at QD256. The wide performance variability again penalizes the average, while the Optimus delivers an average of 23,736 IOPS at QD256.
The Toshiba PX02SMF080 delivers great performance that is above the Optimus as we mix in more writes, until the heavy 90 and 100% write workloads.
The Toshiba has a large percentage of its 8K write activity, 37.15%, land in the range of 20-40ms. 11.34% of I/O falls into the 10-20ms range, and 12.99% lands at 6-8ms. This latency variation reflects the results in the 8k write performance testing. The entire latency range spans from .1-.2ms to 60-80ms.
Power consumption for the Toshiba averages 6.07 Watts, lower than the Optimus at 7.82 Watts.
The Toshiba averages 2,582 write IOPS per Watt.
128K Sequential Read/Write
The 128K sequential speeds reflect the maximum sequential throughput of the SSD using a realistic file size encountered in an enterprise scenario.
The Toshiba PX02SMF080 averages 823MB/s at QD256 in comparison to the Optimus with 586MB/s at QD256. It is important to note that the Toshiba PX02SMF080 is rated for 900MB/s of sequential read speed. We reached these speeds in FOB conditions, but our logging of read performance is conducted after our write testing is completed to assure steady state performance. During this period, we saw results as high as 843MB/s.
The Toshiba exhibits 128K write performance at an average of 371MB/s, 39MB/s shy of its rated speed of 400 MB/s.
The Toshiba PX02SMF080 has a wide range of variability in our mixed read/write workload testing, but manages to average higher speeds than the Optimus until we get into the 80-100% write workloads.
The Toshiba experiences an odd distribution of latency with 128K access that might point out a bimodal performance aspect. The Toshiba provides 444,137 I/Os (49.99%) at 100-200ms, and 224,746 I/Os (25.3%) at 1-2ms, with latency ranging from .2-.4ms to 10-20ms.
The Toshiba PX02SMF080 continues its miserly power consumption with an average of 5.97 Watts during sequential write testing.
The Toshiba provides 61 write MB/s per Watt.
Database/OLTP and Webserver
This test emulates Database and On-Line Transaction Processing (OLTP) workloads. OLTP is in essence 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 Toshiba PX02SMF080 averages 43,296 IOPS during the measurement window, with more pronounced variability than the Optimus, which averaged 39,979 IOPS.
The Toshiba provides 3,075,310 I/Os (23.75%) at 4-6ms, 2,764,853 I/Os (21.36%) at 2-4ms, and 2,501,336 I/O's (19.32%) at 6-8ms. There is a wide gamut of latency for the Toshiba, ranging from .06-.08 to 20-40ms.
The Toshiba averages 6.31 Watts during the precondition run, compared to the 6.8 Watts from the Optimus.
The Toshiba averages 6,817 IOPS per Watt. The Optimus averages 5,823 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 PX02SMF080 averages 43,080 IOPS at QD256, besting the Optimus which averages 33,805 IOPS at QD256. The Toshiba clearly excels at this type of read-centric workload.
The Toshiba provides 32% of I/Os in the 4-6ms range, 28% of I/Os at 6-8ms, 21% of I/O's at 2-4ms, and 13% of I/O's at 8-10ms.
The Toshiba averages 5.3 Watts during the measurement window.
The Toshiba averages 43,800 IOPS per Watt, above the Optimus with 5,579 IOPS per Watt.
Fileserver and Emailserver
The File Server profile represents typical file server workloads. This profile tests a wide variety of different file sizes simultaneously, with an 80% read and 20% write distribution.
The Toshiba PX02SMF080 averages 64,151 IOPS at QD256, well above the 45,736 IOPS mustered by the Optimus.
The Toshiba provides 8,715,777 I/Os (45.42%) at 2-4ms, and 8,185,639 I/Os (42.6%) in the 4-6ms range.
The PX02SMF080 averages 5.61 Watts in steady state very close to the Optimus's average of 5.74 Watts.
The Toshiba averages 11,451 IOPS per Watt, well above the Optimus average of 7,883 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 Toshiba exhibits solid performance in this write-intensive test, with 30,713 IOPS at QD256.
The Toshiba provides 31.18% of I/O's at 10-20ms, 13.75% of I/O's in the 6-8ms range, 11.02% at 1-2ms, and 10.63% at 8-10ms.
The Toshiba averages 6.23 Watts during the measurement window.
The Toshiba averaged 4,903 IOPS per Watt, in comparison to the 4,378 IOPS offered by the Optimus.
The Toshiba PX02SMF080 leverages the dual port SAS 12Gb/s interface to provide the fastest read speeds we have witnessed from a single 2.5" SSD. The PX02SMF080 also brings a mix of enterprise-class features that separates it from the pack in terms of reliability and data security.
Dual port SAS connectivity enables multipath and failover, creating a Tier-0 storage solution deployable into mission critical environments. Enterprise class data security is provided by SANITIZE Cryptographic Erase, TCG SSC protocols and 256-bit AES encryption. The addition of power loss protection and enhanced layered ECC with Toshiba's proprietary QSBC (Quadruple Swing-By Code) secures user data in the event of data corruption or power loss.
A wide range of capacities, up to 1.6TB in a 2.5-inch form factor, provides a scalable solution for end users. A healthy helping of endurance (10 DWPD) provides a robust solution that will survive the rigors of life in the datacenter.
The Toshiba PX02SMx series and its Toshiba/Marvell TC58NC9036GTC controller with 24nm eMLC NAND flaunted some of the fastest read speeds from a single 2.5" device we have witnessed, but there was some significant performance variability in heavy write workloads. We were also unable to reach the maximum sequential speed in steady state, falling short by roughly 40MB/s. This could be the result of the relatively new firmware and hardware combination. In the past, we have witnessed enterprise solid state solutions mature significantly with time in the channel.
When we move into the mixed read/write workloads, the Toshiba PX02SMx series really comes into its own. The SSD performed well in our OLTP, Fileserver and Email server workloads. In read centric workloads, the Toshiba delivered dominating performance with up to 124,000 IOPS available in 4K random read testing.
The Toshiba drive also proved to exhibit miserly power consumption figures with low power draw in all of our workloads. Dual port SAS solutions tend to be a bit power hungry, it is an expected tradeoff for the enhanced capabilities, but the PX02SMF080 defied that expectation. The low power consumption was not reflected entirely in our IOPS to Watts testing, we focus on Write IOPS to Watts. The Toshiba PX02SMF080 is spec'd for 13,400 read IOPS per watt. In our 4K random read workload at QD256 the Toshiba delivered an amazing Read IOPS to Watts ratio of 17,142 IOPS.
We utilized the SMART Optimus as our comparison SSD due to its excellent dual-port performance characteristics. This was a tough task for the 6Gb/s Optimus to stand up to an SSD featuring a faster interface, but the Optimus held up well to the competition. We commend SMART Storage Systems for the performance offered by the Optimus in our testing.
The broader implications of 12Gb/s SAS's enhanced density and new topologies created by longer cabling and management features will take time to penetrate into the datacenter, but storage devices like the Toshiba PX02SMF080 will hasten this transition with their offering of radically improved speed and density. The addition of a five year warranty from a NAND flash fabricator guarantying the longevity and endurance of the SSD, along with bleeding edge speed and connectivity, combine to make the Toshiba PX02SMF series of SSD's a recipient of the TweakTown's Best Performance Award.
Look to these pages soon for a RAID evaluation of the Toshiba PX02SMF080.
PRICING: You can find products similar to this one for sale below.
United States: Find other tech and computer products like this over at Amazon.com
United Kingdom: Find other tech and computer products like this over at Amazon.co.uk
Australia: Find other tech and computer products like this over at Amazon.com.au
Canada: Find other tech and computer products like this over at Amazon.ca
Deutschland: Finde andere Technik- und Computerprodukte wie dieses auf Amazon.de