It did not take long for Samsung to release two new form factors of their EVO series, the mSATA and the M.2. While the mSATA version was expected, the M.2 was a pleasant surprise since that format is favored over the mSATA.
There will be no new features nor ground breaking innovations, both drives are built with the same components as the 2.5in version but housed on a smaller PCB. What it means is, expect the same great performance as the 2.5in version but with a smaller footprint.
The mSATA version is available in 120GB, 250GB, 500GB and 1TB (1000GB) while the M.2. form factor ranges from 120GB, 250GB and 500GB. 5 years limited warranty is now the norm for the EVO line. Only the PRO version offers a 10 years limited warranty.
Usage Application | Client PCs | |
Capacity | 128GB, 256GB, 512GB, 1TB(1000GB) | |
Dimensions (LxWxH) | (29.85±0.15) x (50.80±0.15) x Max 3.85 (mm) | |
Interface | SATA 6Gb/s (backward compatible) | |
Form Factor | mSATA | |
Controller | 120/250/500GB: Samsung MGX Controller 1TB: Samsung MEX Controller |
|
NAND Flash Memory | Samsung 32 layer 3D V-NAND | |
DRAM Cache Memory | 512MB (120GB/250GB/500GB) or 1GB (1TB) LPDDR2 | |
Performance* | Sequential Read: | Max. 540 MB/s |
Sequential Write**: | Max. 520 MB/s | |
4KB Random Read (QD1): | Max. 10K IOPS | |
4KB Random Write (QD1): | Max. 40K IOPS | |
4KB Random Read (QD32): | Max. 97K(250GB/500GB/1TB) Max. 94K(120GB) |
|
4KB Random Write (QD32): | Max. 88K | |
Weight | Max. 8.5g (1TB) | |
Power Consumption | Active Read/Write (Average): Max. 3.7W (1TB)/Max. 4.4W(1TB) | |
Idle: Max. 50mW | ||
Device Sleep: 2mW(120/250/500GB), 4mW(1TB) |
Usage Application | Client PCs | |
Capacity | 128GB, 256GB, 512GB | |
Dimensions (LxWxH) | Max 80.15 x Max 22.15 x Max 0.8 (mm) | |
Interface | SATA 6Gb/s (backward compatible) | |
Form Factor | M.2 (2280) | |
Controller | Samsung MGX Controller | |
NAND Flash Memory | Samsung 32 layer 3D V-NAND | |
DRAM Cache Memory | 512MB (120GB/250GB/500GB) | |
Performance* | Sequential Read: | Max. 540 MB/s |
Sequential Write**: | Max. 500 MB/s | |
4KB Random Read (QD1): | Max. 10K IOPS | |
4KB Random Write (QD1): | Max. 40K IOPS | |
4KB Random Read (QD32): | Max. 97K | |
4KB Random Write (QD32): | Max. 89K | |
Weight | Max. 7g (500GB) | |
Power Consumption | Active Read/Write (Average): Max. 2.4W(500GB) | |
Idle: Max. 50mW | ||
Device Sleep: 2mW |
TRIM Support | Yes (Requires OS Support) | |
Garbage Collection | Yes | |
S.M.A.R.T | Yes | |
Data Security | AES 256-bit Full Disk Encryption (FDE) | |
TCG/Opal V2.0, Encrypted Drive (IEEE1667) | ||
Reliability | MTBF: 1.5 million hours | |
TBW | 120/250GB: 75 TBW 500GB/1TB: 150 TBW |
|
Temperature | Operating: | 0°C to 70°C |
Non-Operating: | -40°C to 85°C | |
Humidity | 5% to 95%, non-condensing | |
Vibration | Non-Operating: | 20~2000Hz, 20G |
Shock | Non-Operating: | 1500G, duration 0.5m Sec, 3 axis |
Warranty | 5 years limited |
Description | Capacity (GB) | Model | $/GB |
850 EVO Series mSATA | 120 | $79.99 | $0.67 |
250 | $129.99 | $0.52 | |
500 | $229.99 | $0.46 | |
1000 | $449.99 | $0.45 | |
850 EVO Series M.2 | 120 | $79.99 | $0.67 |
250 | $129.99 | $0.52 | |
500 | $229.99 | $0.46 |
The components and features of both drives are the same as the 2.5in big brother. The only real difference is the form factor. I am going to summarize the internals and features.
The MGX is a dual core controller, which would be a “downgrade” compared to the three core ARM, MEX. Although per Samsung, the MGX has been optimized for low capacity model and sequential read/write and random low level QD. Which are the most relevant I/O type regarding user experience.
Instead of shrinking, V-NAND stacks vertical layers. By going vertical, there are now more real estates for the cells in size and in between. Which translates into more room for voltage state changes, less interference and lower programing voltage. The benefits are, less retries due to interferences, lower power consumption. The benefits are, less retries due to interferences, lower power consumption, increase NAND endurance and overall I/O performance
TLC or MLC 3 bits per cell really take advantage of the V-NAND since there are more bits cells compared to the MLC (2 bits per cell). Beside the performance increased, the endurance rating was greatly improved allowing Samsung to offer a 5 year warranty compared to the 3 years for the previous EVO generation.
In my opinion, TurboWrite is very ingenious solution to provide SLC performance.
By reserving a small portion of the V-NAND as emulated SLC, that space acts as a buffer, all I/O writes perform at SLC speed. As long as the writes stay under the buffer size (buffer sizes below), write I/O should perform as SLC speed. Once the buffer is full, subsequent writes fall back to TLC performance.
Outside of very specific software, it is unlikely that a home computer usage exceeds 3GB of continuous I/O writes. In most cases, users would be able to take advantage of the SLC speed at the TLC price.
It is unfortunate that the Magician 4.6 was not available for this review. Without it, there was no RAPID feature to test. Based on the previous version, the RAPID technology uses the computer host SDRAM as an extra level of cache. The cache size is dynamic, 25% of the RAM up to 4GB, split 50/50 between read and write. The most noticeable about the write cache is it focuses on small random I/O by collating the data and writes it back in larger blocks. The read cache is persistent, meaning a copy of the data map is written to disk every so often.
Once the Magician 4.6 is released and if there is something worthwhile to mention, I would update this article. I am still hopeful for some customized options to manage the cache.
Dynamic Thermal Guard is great feature especially when we are dealing with mobile device where air flow is premium. The Dynamic Thermal Guard throttles the controller clock if the temperature reach a certain threshold.
“Samsung Data Migration” is a convenient disk cloning software by Clonix. The utility will always detect the OS drive as the source, which is a good failsafe. It only works if at least one SSD is a Samsung. At this time, it is Windows only. The latest version 2.7 puts a new restriction, it only recognized Samsung SSD as the target drive, meaning I can not chose a non-samsung SSD drive as a target. The previous version would only care if at least one drive was a Samsung product.
It is unfortunate that the Magician4.6 was not available. There are quite a few features that require the latest version for implementation.
I went through most of the popular benchmark tools, AS SSD, CrystalDiskMark, ATTO, IoMeter, Anvil’s Storage Utility v1.1.0 and PCMark Vantage. But I also used performance monitoring tools such as DiskMon and hIOmon, primarily to validate the tests. Instead of posting chart after chart, I believe, as a consumer, what is important is how the product fits the needs and not chasing after uber high numbers which are only attainable during benchmarking. For this review, I narrowed it down to Anvil’s Storage Utility, PC Mark Vantage Licensed Pro version, CrystalMark and PCMark8.
Drive conditioning: The SSDs were prepped with Windows 7 (from an image), filled with about 120GB of data total and benchmarks were run from the tested unit acting as the OS drive.
Steady state: This state occurred overtime when the drive went through enough write cycles, or to be more specific program/erase (P/E) cycles, that write performances were consistent or stable. It may take a few weeks before the SSD reaches it, depending on the computing usage, but it can be accelerated using IoMeter.
In summary, Steady State is: Written Data = User capacity x 2, at least.
Benchmark Workstation Main Components | |
CPU | Intel Core i3-2120 Processor @3.30GHz |
Motherboard | ASUSTeK Computer INC. P8Z77-V LX |
RAM | 8,192 MB (4,096 x 2) PNY Optima DDR3 – PC3-10666 |
GPU | NVIDIA GeForce GTS 450 |
OS | Windows 7 Pro 64bits |
Storage Driver | iaStorA 12.9.0.1001 |
OS Hard Drive | The reviewed SSD unit |
mSATA Adapter | mSATA To 2.5in SATA Adapter – SDADA40077 |
M.2 Adapter | ZTC Lightning Card M.2 NGFF SSD – ZTC-EX001 |
Keep in mind that unlike synthetic benchmarks which perform only one specific operation at the time for a predetermined duration, seq read, then seq write then random read, and so on and so forth, real world usage paints a different picture. All four access types can occur at any time, and different transfer rates and different (I/O access) percentages. For instance, a storage subsystem on a streaming server would mostly see high seq read I/O, large block reads, with very little to none write. Looking at a database server without blob data type, we would probably see 75% random read, 20% random write and 5% random and seq write. I could either guesstimate the different ratios or figure a method to define a more accurate I/O usage baseline.
While it is entertaining to run a bunch of benchmarking tools, expecting huge numbers, the purpose of testing the units is to get a good look at how they perform under realistic desktop usage pattern. That is why I picked PCMark Vantage suite as my usage pattern. By capturing and analyzing I/O during the PCVM run, disk operations are breakdown to percentage read vs. write, random vs. sequential, queue depth and average file transfer size.
With that information, benchmarking makes more sense since all the numbers do not carry the same importance, thus some results are more valuable than others.
In summary, I/O pattern defines what I need from the device vs. what can the device do overall.
The I/O baseline process was explained in the Intel 525 mSATA review.
From the numbers, I rated the I/O usage by activity as follow: Random Read > Random Write > Seq Read > Seq Write and average file size is 128K.
To cover Queue Depth, I used hIOmon during the PC Vantage full run. There is a trial version for a week, which is enough time to build the baseline. Based on the chart below, it is obvious that a benchmark score from a QD 16 (or more) does not carry the same weight as a score from a QD 1.
Overall, synthetic benchmarks shows very little difference in performance between the three form factors. If anything, the mSATA and M.2. take a slight lead compared the 2.5in. I would attribute that to a newest firmware.
Traced benchmarks revealed little to no difference between all Samsung version. As long as your computing usage is similar to PC Mark 8 storage, picking an SSD will come down to the price and features.
The appeal for mobile device or small PC format owners is obvious, both formats offer high capacity, desktop like performance, encryption and low power consumption. The whole package is backed by a 5 years limited warranty. Who say one can not have the cake and eat it too?
"WHO SAY ONE CAN NOT HAVE THE CAKE AND EAT IT TOO?"In this day and age, potential owner should aim for the 500GB in capacity. Priced at $0.46 per GB, the 500GB would give the buyer the best ratio performance/price/capacity.
If you are in the market for either drive, you can’t go wrong with one of these.