Test Setup, Benchmarks, And Methodology
|Processor||Intel Core i7-3960X (Sandy Bridge-E), 32 nm, 3.3 GHz, LGA 2011, 15 MB Shared L3, Turbo Boost Enabled|
|Motherboard||Intel DX79SI, X79 Express|
|Memory||G.Skill Ripjaws Z-Series (4 x 4 GB) DDR3-1600 @ DDR3-1600, 1.5 V|
|System Drive||Intel SSD 320 160 GB SATA 3Gb/s|
|Tested Drives||Micron P400m, Firmware: 0200|
|Graphics||AMD FirePro V4800 1 GB|
|Power Supply||OCZ ModXStream Pro 700 W|
|System Software and Drivers|
|Operating System||Windows 7 x64 Ultimate|
|Driver||Graphics: ATI 8.883|
|Iometer v1.1.0||4 Workers, 4 KB Random: LBA=Full, Span Varying Queue Depths|
|ATTO||v2.4.7, 2 GB, QD=4|
|Custom||C++, 8 MB Sequential, QD=4|
|Enterprise Testing: Iometer Workloads||Read||Write||512 Bytes||1 KB||2 KB||4 KB||8 KB||16 KB||32 KB||64 KB||128 KB||512 KB|
The Storage Networking Industry Association (SNIA), a working group made up of SSD, flash, and controller vendors, has produced a testing procedure that attempts to control as many of the variables inherent to SSDs as possible. SNIA’s Solid State Storage Performance Test Specification (SSS PTS) is a great resource for enterprise SSD testing. The procedure does not define what tests should be run, but rather the way in which they are run. This workflow is broken down into four parts:
- Purge: Purging puts the drive at a known starting point. For SSDs, this normally means Secure Erase.
- Workload-Independent Preconditioning: A prescribed workload that is unrelated to the test workload.
- Workload-Based Preconditioning: The actual test workload (4 KB random, 128 KB sequential, and so on), which pushes the drive towards a steady state.
- Steady State: The point at which the drive’s performance is no longer changing for the variable being tracked.
These steps are critical when testing SSDs. It’s incredibly easy to not fully condition the drive and still observe out-of-box behavior, which may lead one to think that it’s steady-state. These steps are also important when going between random and sequential writes.
For all performance tests in this review, the SSS PTS was followed to ensure accurate and repeatable results.
All tests employ random data, when available. Micron's P400m does not perform any data compression prior to writing, so there is no difference in performance-based data patterns.
Seriously though, I wonder why Tom's doesn't run one of their basic "real-world" tests used on consumer SSDs (such as Tom's 7zip test) on one of these professional SSDs just so that we can get an idea how they compare to the consumer-level stuff.
In particular, Tom's always says that comparing one SSD to another is nearly moot point when you consider the magnitude of improvement an SSD has over a traditional HDD; it would be nice to know if these pro-level SSDs are of a similar magnitude of improvement over consumer SSDs or whether the difference is actually less.
The difference is much, much less in terms of performance difference. Tom's has told us this time and time again.
In desktop loads, very very less difference.
In server loads, huge difference. Plus, these server SSD's wil maintain high speeds even after large amounts of data is continuously being written.
How would this compare to a true SLC SSD ?
It'd probably still be inferior overall and not even be cheaper at that point. Over-provisioning is only good for mitigating MLC's disadvantages over SLC AFAIK, not replacing SLC.
Even in server workloads, there are many desktop drives where the performance difference is still not great. For example, Vector is right up there at the tops of the charts with Samsung 840 Pro in performance and many cheaper alternatives are often not far behind in performance. Endurance is another matter, but that wasn't the question. If you want a seriously significant performance difference like with HDDs versus cheap consumer SSDs, you have to consider SSD RAID and/or extreme PCIe storage.
Also, many consumer drives have no trouble keeping performance over time with lots of data written. That's also not an enterprise-only feature.
Some back of the napkin calculations - If you have a consumer 512GB MLC SSD with no over-provisioning, and the MLC was high grade consumer, you could expect 5K P/E cycles. In order to mimic the P400m, which is 35K P/E cycles, the usable space would be ~75GB. By dong that, you would be paying roughly $7/GB for usable storage(assuming you paid $512 for your consumer drive). For SLC, you are looking at 17GB of usable space and $30/GB.
No matter how you look at it, consumer grade MLC will never get close to eMLC and SLC in terms of write endurance, unless the price goes down by orders of magnitude compared to eMLC and SLC.
You are correct. Many drives, such as the 840 PRO, would perform great on the enterprise performance tests. There are also a lot of consumer drives that would have problems over time if they aren't allowed to TRIM every so often. We limit the scope of our testing because the main use cases for these drives are enterprise. Will some companies use them in workstations, absolutely. The same can be said for high-end RAID cards too. Nearly every consumer drive would perform very poorly when you take into account write endurance and other enterprise features.
"These hard drives are really cool...you're not even gonna believe it..."