The latest drive to be added to Kingston's range of data centre drives is the DC1000B, an NVMe drive aimed at a very small but rapidly growing market segment of boot drives for servers. Built around a Phison controller and 64-layer 3D TLC NAND, the DC1000B also features on-board power loss protection. The read centric DC1000B product line consists of just two models, a 240GB drive and the 480GB unit we are reviewing here. At the heart of the drive is a second-generation Phison PCIe Gen3 x4 NVMe controller, the PS5012-E12, which looks after 64-layer 3D TLC NAND. The PS5012-E12 is Phison’s second-generation PCIe Gen3 x4 NVMe controller. The 8-channel controller is built on a TSMC 28nm process and has been designed to work with TLC and QLC NAND technologies with support for Phison’s SmartECC and the latest LDPC (Low-Density Parity Check) error correction as well as AES256, TCG OPAL and TCG Pyrite hardware encryption support. With 8 NAND channels and 32 NAND chip enable lines, the maximum amount of NAND the controller can support is 8TB. Kingston quotes Sequential performance for the 480GB drive as up to 3,200MB/s for reads and up to 565MB/s for writes. The 240GB makes do with up to 2,200MB/s for reads and 290MB/s for writes. 4K random performance for the 480GB drive is quoted as up to 205,000 IOPS for reads and up to 20,000 IOPS for writes. The 240GB drive is rated as up to 111,000 IOPS and 12,000 IOPS for reads and writes respectively. Power consumption wise, the 480GB drive is rated as 1.90W idle, 1.90W and 1.74W for average read and writes respectively. Maximum consumption is listed as 4.88W for reads and 1.81W for writes. As the drive is a fully-fledged data centre/enterprise drive it comes with PLP (Power Loss Protection) hardware installed on the drive and it supports SED (Self-Encrypting Drive) via AES 256-bit encryption. Endurance wise, the 480GB is rated as 475TBW or around 0.5 DWPD for the length of the 5-year warranty Kingston backs the drive with. Physical Specifications: Usable Capacities: 480GB. NAND Components: 64-layer 3D TLC NAND. NAND Controller: Phison PS5012-E12. Cache: 1TB. Interface: PCIe NVMe Gen3 x4. Form Factor: M.2 2280. Dimensions: 80 x 22 x 3.8mm. Drive Weight: 9g. Firmware Version: ECEK22.3 The DC1000B ships in a blister pack with the drive’s capacity clearly labelled on the front while the rear has multilingual marketing and warranty notes on it. The DC1000B is built on a dual-sided M.2 2280 format with components on both sides of the PCB. Under the product sticker sit two packages of 64-layer 3D TLC NAND, the Phison PS5012-E12 PCIe Gen3 x4 NVMe 8 channel controller and a DRAM cache IC. Also present are a row of four PLP (Power Loss Protection) modules. The other side of the PCB has two more NAND packages, another cache IC and three more power protection modules. Although PLP is common on longer enterprise M.2 drives, it's a rare thing to see on a 2280 format drive. Kingston’s SSD management software utility is simply called SSD Manager. With it, you can monitor the health of the drive and how it’s being used, check the drive’s SMART data (including reliability tracking, usage statistics, life remaining, wear levelling and temperature) and update the firmware. For testing, the drives are all wiped and reset to factory settings by HDDerase V4. We try to use free or easily available programs and some real-world testing so you can compare our findings against your own system. This is a good way to measure potential upgrade benefits. Main system: Intel Core i7-7700K with 16GB of DDR4-3200 RAM, Sapphire R9 390 Nitro and an Asus Prime Z270-A motherboard. Other drives Intel Optane SSD DC P4800X 750GB Intel SSD DCP4510 8TB Kingston DCP1000 1.2TB (RAID 0 & 1) Kingston KC1000 960GB Software: Atto Disk Benchmark CrystalMark 7 IOMeter. Futuremark PC Mark 8 All our results were achieved by running each test five times with every configuration this ensures that any glitches are removed from the results. Trim is confirmed as running by typing fsutil behavior query disabledeletenotify into the command line. A response of disabledeletenotify =0 confirms TRIM is active. CrystalDiskMark is a useful benchmark to measure theoretical performance levels of hard drives and SSD’s. We are using v7.0. The latest version 7 of CrystalDiskMark includes a couple of profiles that can be used for testing along with the standard tests, Peak Performance and Real World. In both the standard and Peak Performance tests, the drive passed the official read Sequential ratings of 3,200MB/s. Using the Real World profile the Sequential read performance dropped to 2,852MB/s and 1,880MB/s when using compressible and incompressible data respectively. The only time the Sequential write test results fell below the official 565MB/s was when the drive was being tested in the default mode using incompressible data, where the performance dropped to 531MB/s, the rest of the time the test results were around 30MB/s faster than the official figure. Kingston's 4K random read figure for the drive is up to 205,000 IOPS when the drive is in a Steady State. Fresh out of the box the best read figure we got from the CrystalDiskMark benchmark was 357,958 IOPS at a QD of 8 using compressible data. The official 4K random write figure for the 480GB drive is up 20,000 IOPS (in Steady State). The best figure we could see from our CrystalDiskMark queue depth testing out of the box was 147,476 IOPS using four threads at a QD of 1. Looking at the two sets of CrystalDiskMark result screens it appears that the Phison PS5012 controller is much more efficient when reading compressible data at certain queue depths. The ATTO Disk Benchmark performance measurement tool is compatible with Microsoft Windows. Measure your storage systems performance with various transfer sizes and test lengths for reads and writes. Several options are available to customize your performance measurement including queue depth, overlapped I/O and even a comparison mode with the option to run continuously. Use ATTO Disk Benchmark to test any manufacturers RAID controllers, storage controllers, host adapters, hard drives and SSD drives and notice that ATTO products will consistently provide the highest level of performance to your storage. Kingston quote Sequential figures for the drive as up to 3,200MB/s for reads and 565MB/s for writes. We could confirm both figures using the ATTO benchmark with the review drive producing a read figure of 3,190MB/s with writes at 571MB/s. In our 128KB Sequential tests, we got result figures that bettered the official numbers for both reads and writes. The read result of 3,507MB/s is some 307MB/s better than the official figure while the tested write figure of 638MB/s is 73MB/s faster than the official figure. Kingston rate the 4K random read performance of the 480GB version of the DC1000B as up to 205,000 IOPS (with the drive in a Steady State). With our four-threaded tests we were able to squeeze a bit more out of the drive, 211,074 IOPS, (at a QD of 32) when fresh out of the box. Kingston quote a Steady State 4K random write figure of up to 20,000 IOPS for the drive. When tested fresh out of the box, we got a peak figure of 138,223 IOPS at a QD of 1 with our four-threaded tests. The drive performs very well in our 4K 70/30 read/write test, with the performance climbing steadily and smoothly through the tested queue depths. In our read throughput test, the drive peaked at 2,121.17MB/s at the 4MB block mark, somewhat short of the maximum 3,200MB/s quoted by Kingston. Writes peaked at the 64KB block size at 724MB/s, over 150MB/s faster than the official figure of 565MB/s before dropping back to finish the test run at 561MB/s. PCMark 8’s Standard Storage test also saves a large amount of performance data. The default test runs through the test suite of 10 applications three times. Here we show the total bandwidth performance for each of the individual test suites for the third and final benchmark run. The read optimised DC1000B offers fairly decent performance in PCMark 8's Standard Storage test but it's bandwidth figures are quite a way off what we would see from a more balanced (in terms of read and writes) NVMe drive, especially the two Adobe Photoshop tests. For the long term performance stability test, we set the drive up to run a 20-minute 4K random test with a 30% write, 70% read split, at a Queue Depth of 256 over the entire disk. The DC1000B 480GB averaged 68,822 IOPS for the test with a performance stability of 56%. As the DC1000B has been primarily designed as a boot drive we thought we would time how fast the drive boots from a cold start into a ready to use, basic installation of Windows Server 16. We compared it to a SATA based 2.5in SSD and a fast mechanical hard drive. As you can see the DC1000 boots in half the time of a standard HDD and two seconds faster than a SATA interfaced SSD. To test real life performance of a drive we use a mix of folder/file types and by using the FastCopy utility (which gives a time as well as MB/s result) we record the performance of drive reading from & writing to a 256GB Samsung SSD850 PRO. 100GB data file. 60GB iso image. 60GB Steam folder – 29,521 files. 50GB File folder – 28,523 files. 21GB 8K Movie demos. 11GB 4K Raw Movie Clips (8 MP4V files). 4.25GB 3D Printer File Folder (mostly .STL). 1GB AutoCAD File Folder (.dwg and .dxf). The Kingston DC1000B had no problems dealing with our real-life file transfer tests. The drive is much more efficient at handling larger file sizes than the smaller files found in the 60GB Steam and 50GB file folders as can be seen by the large difference in transfer rates. To get a measure of how much faster PCIe NVMe drives are than standard SATA SSD's we use the same files but transfer to and from a 512GB Toshiba OCZ RD400. Taking the SATA SSD out of the picture and switching over to transfer the files to and from the NVMe drive shows a huge leap in read performance with a much smaller gain in writes when dealing with larger file sizes. Although Kingston's DC1000B drive is has been optimised to operate as a low-cost boot drive for servers it can also be used in a data logging role or as a local cache drive. As more and more server motherboards come into the market equipped with NVMe M.2 slots it makes sense to use them to house the boot drive rather than take away one of the main storage drives to perform the boot function. For the DC1000B, Kingston has gone down the Phison/3D TLC controller and NAND combination route. The controller is the 8-channel PS5012-E12, Phison’s second-generation PCIe Gen3 x4 NVMe IC while the NAND is 64-layer 3D TLC. The 480GB drive uses four NAND packages, two on each side of the PCB. The DC1000B is heavily focused on read performance which becomes obvious when looking at the official performance ratings for the drive. When it comes to Sequential performance, Kingston rate the 480GB drive at up to 3,200MB/s for Sequential reads with writes down at up to 565MB/s. Using the ATTO benchmark we could confirm those official ratings with the tested drive producing a read figure of 3,190MB/s with writes at 571MB/s. Random performance is quoted as up to 205,000 IOPS for reads and 20,000 IOPS for writes, both figures are from a drive in a Steady State. Tested fresh out of the box with our four threaded tests, we got a read figure of 211,074 IOPS (at a QD of 32). Peak tested write performance came at a QD of 1 at 138,223 IOPS. It comes as no surprise to find that a drive aimed at the data centre/enterprise market segment should come with onboard power loss protection (PLP) but what is surprising is to find it on an M.2 2280 format drive. Usually, PLP is found on longer M.2 format drives for servers, but Kingston has found a way to add to a shorter PCB. We found the 480GB Kingston DC1000B on Span.com for £168 (inc VAT) HERE Discuss on our Facebook page HERE. Pros Sequential read performance. Endurance. Power loss protection. Cons Pricey. KitGuru says: Kingston's latest NVMe SSD is aimed at a rapidly growing segment, namely server boot drives. With more server motherboards entering the market space with one or more M.2 NVMe slots, it means that you no longer have to use up one of the servers storage drives to perform boot duties.
KitGuru KitGuru.net – Tech News | Hardware News | Hardware Reviews | IOS | Mobile | Gaming | Graphics Cards
