Intel Solid-State Drive DC S3700 Review

3 thumbs up

Our Test Methodologies: Under each test condition, the Solid State Drives tested here were installed as secondary volumes in our testbed, with a standard spinning hard disk for the OS and benchmark installations. Out testbed's motherboard was updated with the latest BIOS available as of press time and AHCI (or RAID) mode was enabled. The SSDs were secure erased and left blank without partitions wherever possible, unless a test required them to be partitioned and formatted, as was the case with our ATTO, PCMark 7, and CrystalDiskMark benchmark tests. Windows firewall, automatic updates and screen savers were all disabled before testing. In all test runs, we rebooted the system, ensured all temp and prefetch data was purged, and waited several minutes for drive activity to settle and for the system to reach an idle state before invoking a test.

HotHardware Test System
Intel Core i7 and SSD Powered

Processor -

Motherboard -


Video Card -

Memory -

Audio -

Hard Drives -

 

Hardware Used:
Intel Core i7-2600K

Asus P8Z6-V Pro
(Z68 Chipset, AHCI Enabled)

NVIDIA GeForce GTX 285

4GB Kingston DDR3-1600

Integrated on board

WD Raptor 150GB (OS Drive)
OCZ Vertex 3 Pro(480GB)
Corsair Force GT (240GB)
Intel SSD 710 (200GB)
Intel SSD 520 (240GB)
Intel DC S3700 (200GB, 800GB)

OS -
Chipset Drivers -
DirectX -

Video Drivers
-


Relevant Software:
Windows 7 Ultimate SP1 x64
Intel 9.2.0.1030, iRST 10.5.1027
DirectX 11

NVIDIA GeForce 275.33

Benchmarks Used:
IOMeter 1.1.0 RC
HD Tune v4.61
ATTO v2.47
AS SSD
CrystalDiskMark v3.01 x64
PCMark 7
SiSoftware Sandra 2011

IOMeter
I/O Subsystem Measurement Tool

As we've noted in previous SSD articles, though IOMeter is clearly a well-respected industry standard drive benchmark, we're not completely comfortable with it for testing SSDs. The fact of the matter is, though our actual results with IOMeter appear to scale properly, it is debatable whether or not certain access patterns, as they are presented to and measured on an SSD, actually provide a valid example of real-world performance for the average end user. That said, we do think IOMeter is a reliable gauge for relative available throughput within a given storage solution. In addition there are certain higher-end workloads you can place on a drive with IOMeter, that you can't with most other storage benchmark tools available currently.

In the following tables, we're showing two sets of access patterns; our custom Workstation pattern, with an 8K transfer size, 80% reads (20% writes) and 80% random (20% sequential) access and IOMeter's default access pattern of 2K transfers, 67% reads (34% writes) and 100% random access.

 

We've got a bunch of numbers to share on the Intel DC S3700 series solid state drives. We tested both the 200GB and 800GB models in single drive and dual-drive RAID 0 configurations, and compared them to a number of other Intel built or enterprise-class drives throughout.

Here in our IOMeter tests, you can see the Intel DC S3700 offers flat, consistent performance across the board, regardless of the queue depth or access pattern. The single-drive configurations trailed the other drives we tested in total IOPS, but the RAID configurations fared much better, though the still trailed a couple of other setups.

Transfer speed for the Intel DC S3700 was also lower than competing offerings, especially those based on LSI SandForce controller technology.
 

Article Index:

0
+ -

I like that you call it cheap in comparison to other enterprise drives, but then list it as "realtively expensive" under the "nots". That's like calling a porche expensive when comparing it to the price of a kia.

But yeah, I agree it's too expensive for us still living in mud huts.

0
+ -

Well, it's relatively expensive in light of all SSDs, but it's not for its target market.

0
+ -

Marco, re: "the Intel DC S3700 offers flat, consistent performance across the board, regardless of the queue depth or access pattern".

This is a clear sign of a design defect. When performance does not improve as a function of queue depth, it means that the drive is essentially broken, incapable of parallelization of queued I/O request.

That Intel marketeers can turn this around into some bizarre kind of "advantage" is amusing, at best.

From what I can see, these new "Data Center" class SSDs from Intel are optimized only to increase Intel's profit margins.

Apparently they do this by eliminating the licensing costs of the LSI Sandforce controller, which outperforms the Intel silicon by a factor of 2x-3x.

0
+ -

ClockSpeedy, Keep in mind this Intel 'enterprise' drive is an MLC model which is why its more reasonably priced. The SLC enterprise SSDs are still the way to go if speed and reliability are the key factors.

Login or Register to Comment
Post a Comment
Username:   Password: