The award winning CloudSpeed 500 from Smart Storage Systems offers up admirable endurance specifications and provides enterprise-class features at a price point competitive with consumer hardware. This addresses the gap between heavy-workload SSDs and client SSD products.
The 6Gb/s CloudSpeed 500 delivers 500MB/s in sequential read/write speed and 60,000/20,000 random read/write IOPS. The SSD is available in 120, 240, and 480GB capacities. Featuring the proven LSI SandForce SF-2581, an enterprise version of the SF-2281 controller, in tandem with SMART Storage Systems proprietary technologies provides a solid platform for the CloudSpeed 500.
The Guardian Technology Suite, covered in more detail on the following page, allows SMART Storage Systems to take typical consumer-grade MLC NAND and extend its lifetime to 15,000 P/E cycles. This is a 3x-5x improvement over normal MLC NAND. This allows the CloudSpeed 500 to withstand 1.2 full random drive writes per day (DWPD). This important specification underlines one of the key advantages of the CloudSpeed 500 over client SSDs.
Pricing is always a big concern, and many times the exorbitant pricing of some high-end enterprise-class SSDs can lead administrators to deploy client SSDs into servers to save on upfront acquisition costs. By deploying relatively cheap client hardware, and simply replacing the SSDs as they wear out, companies can enjoy large cost savings over some enterprise SSDs.
This 'Rip-and-Replace' model provides initial advantages in the pricing department, but there are significant trade-offs in performance and data protection. Consumer SSDs are not designed for the heavy workloads of enterprise environments, and performance falls rapidly in these scenarios.
The CloudSpeed 500 also includes power capacitors not found in consumer hardware, which allow all data to be committed to NAND in the event of a host power-loss issue. SandForce processors also feature R.A.I.S.E. (Redundant Array of Independent Silicon Elements) technology. This baked-in feature provides data redundancy in the event of uncorrectable errors.
R.A.I.S.E., along with full data path protection, provides a higher level of protection for data not offered by consumer products. In concert with ECC, this results in an UBER (Uncorrectable Bit Error Rate) of less than 1 in 10^18 bits read. This is a tremendous 100X improvement over the JEDEC JESD 218 spec for enterprise-grade SSDs.
SMART Storage Systems brings the features and reliability of a tier-one OEM enterprise SSD into a competitive price point that in the past was unattainable for an SSD with these features. Backing the enhanced endurance, performance, and data protection technologies with a five year warranty gives SMART's customers peace of mind that their storage devices will function reliably in demanding scenarios.
CloudSpeed 500 Specifications
The award-winning Guardian Technology Platform is SMARTs proprietary technology suite that consists of several integral components.
Before assembly, SMART subjects all NAND to an extensive re-characterization, which reassigns more accurate P/E ratios to the NAND. Aggregated flash management ensures that there is an even distribution of different classes of NAND present on each SSD to benefit the whole.
NAND is typically programmed for retention and wear metrics during NAND processing and the settings are never altered. Adaptive programming dynamically adjusts flash parameters over the course of the SSDs life, tailoring the NAND settings at the base level. This creates exponential increases in endurance of the underlying NAND.
The third generation host power-loss protection approach protects against unexpected power interruptions. An array of highly reliable discrete Tantalum capacitors provides the power to flush all data in transit to the NAND during loss of power. These Tantalum capacitors are rated for high temperature environments and do not experience degradation over time.
The SSD comes in the standard 2.5" form factor with a 7mm Z-Height. This will allow deployment of the SSD into slim applications. These features are rounded out by AES-128 bit encryption and a five year warranty.
CloudSpeed 500 Internals
The CloudSpeed 500 comes in a 2.5" form factor with a slim 7mm Z-Height. This will allow for deployment into blade servers and dense applications. The SSD also features internal thermal monitoring and enhanced SMART statistics for monitoring the SSD.
The SSD comes in a 2.5" form factor aluminum chassis.
The SandForce SF-2581 controller rests flush against a thermal pad. This allows the SSD to withstand temperatures up to 70C.
The 24nm MLC NAND is rebadged with SMART Storage Systems logo. This NAND has been through a re-characterization process and optimized to provide 15,000 P/E cycles.
There is 256GB of raw NAND onboard the CloudSpeed 500, and 224GB is user addressable.
The third-generation power loss protection is provided by an array of capacitors along the edge of the PCB. We can also observe the additional open pads for more capacitors with the larger capacity version of the CloudSpeed 500. These capacitors provide reliable protection in high-heat environments.
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 take latency distribution into consideration, 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.
The first page of results will provide the 'key' to understanding and interpreting our new test methodology.
4K Random Read/Write
We precondition the CloudSpeed 500 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 light brown 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 provide histograms for further latency granularity below. This downward slope of performance happens very few times in the lifetime of the device, typically during the first few hours of life, 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 CloudSpeed 500 provides 49,000 IOPS at QD256, and the SM843 averages 69,654 IOPS in random 4K read speed at QD256. The SM843 performs well in read-centric environments, and provides much higher performance.
Garbage collection routines are more pronounced in heavy write workloads. This leads to more variability in performance, and the CloudSpeed 500 has a wide range of performance in this test, averaging 25,560 IOPS in 4K write speed at QD256. The Samsung SM843 averages 12,313 IOPS. While the CloudSpeed 500 does suffer from more pronounced variability than the SM843, its performance in a heavy random write workload is higher.
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 CloudSpeed 500 has a reduction in performance from 49,000 IOPS in 4K read to 25,560 IOPS with a pure 4K random write workload.
The SM843 provides steady performance from the 10% to 60% range, but then falls to an average of 13,000 IOPS at a 100% 4k write workload. We can observe that in pure 4K read workloads the SM843 is hard to beat, but once we apply a 10% write workload, it falls quickly.
The CloudSpeed 500 has 3,474,579 I/O's (44.9%) fall into the 4-6ms range, and 4,078,199 I/O's (52.7%) fall into the 10-20ms range.
The CloudSpeed features a tighter latency range than the SM843 in the heavy write workload testing. The SM843 is optimized for read-heavy workloads, so this distribution is not entirely unexpected.
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 CloudSpeed 500 is higher than the SM843, with an average of 3.7 Watts. The SM843 averages 2.79 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 CloudSpeed 500 with an advantage in the 4K random write workload at 6,994 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 CloudSpeed 500 comes in at 30,000 IOPS, but the SM843 manages a tremendous 44,499 IOPS during the QD256 measurement window.[img]20 [/img]
The average 8K random write speed of the SMART Storage Systems CloudSpeed 500 is 16,460 IOPS at QD 256, and the Samsung SM843 averages 6,269 IOPS.
The CloudSpeed has the same slight performance curve with the addition of a write workload. The SM843 leads the pack with the pure 8K random read speed, but performance falls once we begin mixing in a write workload.
The CloudSpeed has very good latency distribution in this test, with 2,142,093 I/O's (43.6%) falling into the 20-40ms range, and 1,754,697 I/O's (35.7%) falling into the 6-8ms range.
Power consumption for the CloudSpeed 500 averages 3.39 Watts, and the SM843 averaged 2.67 Watts.
The CloudSpeed averages 4,792 IOPS per Watt.
128K Sequential Read/Write
The 128K sequential speeds reflect the maximum sequential throughput of the SSD using a realistic file size actually encountered in an enterprise scenario.
Both the CloudSpeed 500 and the SM843 are closely matched in this test, with an average hovering around 530MB/s.
The CloudSpeed falls in sequential write speed to the SM846, with an average of 425MB/s. The SM846 averages an impressive 487MB/s in sequential write speed at QD256.
The CloudSpeed 500 again manages good marks as we mix in a read workload in steady state conditions.
Both SSDs give great latency performance at 128K sequential write, with the SM843 actually providing 100% of its I/O's in the 60ms range. The CloudSpeed 500 provides 982,528 I/O's (93.72%) at 60-80ms, and 65,846 I/O's (6.28%) at 80-1900ms.
The CloudSpeed 500 averages 3.83 Watts, and the SM843 averages 3.65 Watts.
The CloudSpeed 500 gives 113MB/s per Watt, and the SM843 provides 129MB/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 CloudSpeed 500 manages 25,100 IOPS at QD256, and the SM843 averages 13,481 IOPS.
The CloudSpeed 500 provides 3,637,315 I/O's (48%) in the 10-20ms range, 2,246,858 I/O's (29.7%) in the 8-10ms range, and 1,523,348 I/O's (20.1%) in the 6-8ms range.
The CloudSpeed 500 averages 3.1 Watts, in comparison to the SM843 with 2.47 Watts.
The CloudSpeed averages 8,152 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 websites, and thus the end user experience.
The CloudSpeed averages 19,886 IOPS at QD256, and the SM843 averages 27,919 IOPS. This read-heavy workload is well suited to the SM843's strengths.
The CloudSpeed provides 5,555,344 I/O's (93.4%) in the 10-20ms range and 374,791 I/O's (6.3%) in the 8-10ms range.
The CloudSpeed 500 averages 3.41 Watts in the Webserver workload, much higher than the 2.06 Watts of the SM843.
The CloudSpeed 500 provides 5,881 IOPS per Watt in this test, but the SM843 scores an average of 13,100 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 CloudSpeed 500 averages 26,203 IOPS, and the SM843 averages 36,413 IOPS at QD256.
The CloudSpeed 500 provides 4,895,994 I/O's (62%) in the 8-10ms range, and 2,849,878 I/O's (36.3%) in the 10-20 range.
The CloudSpeed 500 averages 3.6 Watts, and the SM843 averages a much lower 1.96 Watts.
The CloudSpeed averages 7,300 IOPS per Watt in steady state, and the SM843 averages 18,000 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 CloudSpeed averages 23,036 IOPS at QD256, and the SM843 averages 11,911 IOPS.
The CloudSpeed 500 provides 3,417,735 I/O's (47.1%) in the 10-20ms range, 2,598,933 I/O's (35.8%) at 8-10ms, and 1,067,774 I/O's (14.7%) at 6-8ms.
The CloudSpeed 500 averages 3.62 Watts in steady state, and the SM843 averages 2.69.
The CloudSpeed averages 6,579 IOPS per Watt, ahead of the SM843.
The simple addition of an SSD will provide an exponential performance increase over any HDD, and we are witnessing an increasing number of administrators utilizing consumer SSDs to realize these gains with minimal investment. This rip-and-replace model does not always provide the expected TCO advantages, and performance and reliability tradeoffs become a concern.
This trend has created a new market segment, with SSD manufacturers jumping in to fill the void between heavy workload enterprise SSDs and client-side products. The CloudSpeed 500 is geared for the value market, and is intended to compete directly with consumer SSDs. The CloudSpeed 500 provides performance and endurance advantages that will be hard to match for typical client-side hardware.
The Samsung SM843 is another SSD intended to compete in this segment, and the match-up between these two drives is close in many aspects. As we move forward with more product evaluations, we will see the test pool for this product segment grow very quickly.
The aggressive pricing of the SM843 also places it within striking range of consumer products. The Samsung SM843 tends to fall into a read-centric performance envelope. We witnessed strong performance in several read-centric workloads, particularly the webserver and fileserver emulations. However, the CloudSpeed 500 features more robust performance characteristics in heavier write workload environments. The consistent performance in mixed read/write workloads is an advantage for those looking for a competitively priced SSD.
The CloudSpeed 500 also offers up host power-fail protection with SMART Storage Systems Power Guard technology. This provides the SSD the ability to write all data down to the NAND in the event of a power failure. This is a feature we commonly see in expensive enterprise SSDs, and is a welcome inclusion at this price point.
The CloudSpeed price range depends upon many factors, so exact pricing figures can fluctuate. The CloudSpeed normal price range is in the $1 to $1.50 range per GB. For such competitive pricing, the CloudSpeed 500 features great performance in heavy write workloads, and the addition of power fail technology is a big advantage over competing devices.
Utilizing the proven LSI SandForce SF-2581 provides a reliable platform with several key data protection features baked-in. R.A.I.S.E. and Guardian Technology combine to provide a level of data security a surprising 100X the JEDEC specification for enterprise-class SSDs.
SMART Storage Systems enjoys the reputation and reliability of a tier-one OEM supplier, and the along with the five year warranty, it provides this durable SSD a clear advantage over consumer products, and many enterprise competitors.
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