Samsung SSD XP941 Review: The PCIe Era Is Here
by Kristian Vättö on May 15, 2014 12:00 PM ESTMac Benchmarks: QuickBench, AJA & Photoshop Installation
Since the XP941 is currently only bootable in Macs, I decided to run some benchmarks with the XP941 inside a Mac Pro. The specs of the Mac Pro are as follows:
| Test Setup | |
| Model | Mac Pro 4.1 (Early 2009) |
| Processor | Intel Xeon W3520 (2.66/2.93GHz, 4/8, 8MB L3) |
| Graphics | NVIDIA GeForce GT120 512MB GDDR3 |
| RAM | 12GB (2x4GB + 2x2GB) DDR3-1066 ECC |
| OS | OS X 10.9.2 |
We would like to thank RamCity for providing us with the Mac Pro, so we were able to run these tests and confirm boot support.
I installed OS X 10.9.2 to all drives and they were the boot drives when benchmarked, just like they would be for most end users. As I mentioned on page one, RamCity actually sent us two 512GB XP941 and I just had to put them in RAID 0 configuration. With a Mac you can easily boot from a software RAID 0 array, so all I had to do was to create a RAID 0 array in Disk Utility and select it as the boot volume. I placed the drives in PCIe slots 2 and 4 to ensure that both drives were getting full PCIe bandwidth and we wouldn't run into bottlenecks there. I picked Intel's 480GB SSD 730 to be the comparison point as it was lying on my table and is among the fastest SATA 6Gbps SSDs in the market. Note that the 2009 Mac Pro only supports SATA 3Gbps, so there's obviously some performance penalty from that as the benchmarks show.
QuickBench
QuickBench is one of the more sophisticated drive benchmark tools for OS X. It's shareware and retails for $15 but compared to the freeware tools available, it's worth it. While QuickBench lacks the option to increase queue depth, it supports various transfer sizes from 4KB to up to 100MB (or more through a custom test). For this test, I just ran the standard test where the IO sizes range from 4KB to xMB. Additionally I ran the extended test, which focuses on very large IOs (20-100MB) in order to get the maximum performance out of the drives. In both cases the tests ran for 10 cycles to ensure sustained results.
The random results don't reveal anything interesting. The RAID 0 array is slightly slower due to the overhead from the software RAID configuration but overall the results make sense when compared with our Iometer scores. Bear in mind that QuickBench only uses queue depth of 1, whereas our Iometer tests are run at queue depth of 3, hence there's a difference that is roughly proportional to the queue depth.
The sequential tests show that the XP941 seems to be slightly slower in the Mac Pro compared to sequential performance in Iometer. In this case both tests are at a queue depth of 1 and should thus be comparable, but it's certainly possible that there are some other differences that cause the slightly slower performance. Either way, we are still looking at much, much higher performance than any drive would provide under the Mac Pro's native SATA 3Gbps interface.
Since QuickBench doesn't allow increasing the queue depth, the only way to increase performance is to scale the transfer size. QuickBench's preset tests allow for up to 100MB IO sizes and I ran the preset that tests from 20MB to 100MB and picked the highest perfoming IO sizes that were 90MB and 20MB in this case. There wasn't all that much variation but these seemed to be the highest performing IO sizes for all three configurations.
Now the XP941 and especially RAID 0 show their teeth. With two XP941s in RAID 0, I was able to reach throughput of nearly 2.5GB/s (!) and half of that with a single drive. Compared to the SSD 730 in the SATA 3Gbps bus, you are getting over four times the performance and to reach the performance of X941 RAID 0 you would need at least ten SSDs in a SATA 3Gbps RAID 0 configuration.
AJA System Test
In addition to QuickBench, I decided to run AJA System Test as it's a freeware tool and quite widely used to test disk performance. It's mainly designed to test the performance of video throughput but as the results are reported in megabytes per second, it works for general IO testing as well. I set the settings to the maximum (4096x2160 10-bit RGB, 16GB file size) to product the results below.
The results are fairly similar to the QuickBench ones but the performance seems to be slightly lower. Then again, this is likely due to the difference in the data the software uses for testing but the speeds are still well over 1GB/s for a single drive and 2GB/s for RAID 0.
Adobe Photoshop CS6 Installation
One of the most common criticism I hear towards our tests is that we don't run any real world tests. I've been playing around with real-time testing a lot lately in order to build a suite of benchmarks that meet our criteria but for this review I decided to run a quick installation benchmark to see what kind of differences can be expected in real world. I grabbed the latest version of Photoshop CS6 trial from Adobe's website and installed it to all three drives while measuring the time with a stopwatch.
Obviously the gains are much smaller in typical real world applications. That's because other bottlenecks come to play, which are absent when only testing IO performance. Still, especially for IO heavy workloads the extra performance is always appreciated even if the gains aren't as substantial as benchmarks show.
Facebook
Twitter
RSS
110 Comments
View All Comments
RamCity - Monday, May 19, 2014 - link
The Lycom M.2 to PCie adapter does a reasonable job with small form factor computers. It has a low-height bracket in addition to the standard height one. Note though, you can't boot the XP941 in any windows PC with one of these adapters as far as we know - it wont show up in the bios as a bootable device. Only in a 2006-2012 Mac Pro under OSX will it be bootable.Rod
Oscarcharliezulu - Monday, May 19, 2014 - link
Nice job anandtech... You detected what country i was reading this from and inserted appropriate text/prices? Great job. Appreciated.kizh - Tuesday, May 20, 2014 - link
still running my sandy 2500k that did 4.8 reliably until I updated bios. I bought a z67 board that had a pcie 3.0 x 16 lane as an early adopter. The only thing worth upgrading on it was the graphics so bought a 780 (on accident thought it was a 780 ti) I spent about 24 hours of my time researching making it work on a early version of uefi and trying multiple set ups. Gonna send it back, its not worth upgrading my whole system for this.Never buying another feature that isn't being used yet, by the time its truly a standard your warranty is over and FU if it got tweaked a little.
Got 2 M2 in raid, no trim ever bothered me. Its fast enough, don't need this. Maybe if a whole system upgrade was as fun as it used to be,
Right now all I see waiting for is a 4k display port standard and some nicer models coming out. I want to throw money at these guys but not for next to nothing,
I'm not rich by any measure but dumped money on tech as a hobbyist.
kizh - Tuesday, May 20, 2014 - link
oh just to clarify I know it would only be pcie 3.0 with ivy, I didn't expect that chip to be such a let down. Still this should have worked at pcie 2.0. Also when I say m2 I'm talking about the model name of my ssd's. not m.2just some clarifications
sfgebrqy - Wednesday, May 21, 2014 - link
I would have bought this in a heart-beat if it weren't for two things:1) Average/bad 4k random read/write
2) Abysmal 4k scaling with QD
The killer feature for this SSD is without a doubt its sequential performance. Compared to any single SSD alone, it blows it out of the water on the benches, whether
practical or synthetic.
The caveat is "single SSD". RAID 0 can always be used to boost sequentials. Maybe not exactly linearly, but close to linearly. Take a look at the review for 2 Samsung 840
Pros in RAID 0 (page 3). I can't link (or even mention the website) it because the comment filter says its spam, so take a guess and use Google. Sequentials are nearly
doubled for both read and write.
The kicker is that this CANNOT be done for random I/O at low QD. Which is to say, if we wanted better random I/O from XP941s, we couldn't RAID them together to do it. In
fact, RAID typically makes random I/O marginally worse at low QD due to overhead. See the aforementioned review (page 4).
RAID does multiply random I/O at high QD, but with XP941 4k QD scaling non-existent to begin with, it's a lost cause.
Even if we did want to RAID XP941s together, I can't seem to find any tests/reviews of more than two at once, let alone a RAID controller capable of putting more than two
M.2 devices in RAID 0. Meanwhile, there are plenty of RAID controllers capable of pushing 16+ SATA3 devices (albeit bottlenecked at x8 PCIe 3.0).
The bottom line is that if I found out that my current SATA3 SSD RAID 0 array wasn't putting out enough sequential, I could just add a few more SSDs to solve the issue. If I
found out that my XP941 array wasn't putting out enough random I/O, there's nothing I could do to fix it.
If you're in the market for single SSDs and your system can handle M.2, then this is the cream of the crop. If you have RAID arrays though, wait until M.2 becomes more
popular and Samsung pushes a new M.2 SSD with better random I/O performance.
skrewler2 - Sunday, May 25, 2014 - link
Sorry late to the party but this part of the article is just ignorant:"Using an SSD as a secondary drive can make sense for e.g. a video professional where the performance can be utilized as a scratch disk, but otherwise the only real use case for an SSD is as a boot drive"
What the hell?
aviv - Wednesday, June 4, 2014 - link
Hey gr8 reviewI got question about the graphs
Its log systemic
Does it means when its lower by 1/10 its just half of the value or
In between its linar
brentpresley - Thursday, July 3, 2014 - link
ASUS just released a new BIOS today (7/3/2014) that allows for M.2 boot support of the XP941.I just installed it and the BIOS can see the XP941 in AHCI mode now.
iBurley - Wednesday, December 3, 2014 - link
If put into the M.2 slot of one of the ASRock Z97 boards mentioned in the review, will this be running at full speed, or would I need an adapter like you tested with to use the full potential of the drive?On ASRock's website, under the port listing for the board, it states "1 M.2 (PCIe Gen2 x2 & SATA, Supports 30mm, 42mm, 60mm, 80mm, 110mm M.2 devices)" but I don't know if that just means the interface or if it would actually go over the PCIe bus instead of SATA.
SERGE 2015 - Tuesday, July 14, 2015 - link
I have AsRock's Z97 Extreme6 and XP941 and it is unfortunately NOT bootable out of the box. As it's been mentioned - once the system boots up using another EFI source, the XP941 becomes visible and can then load whatever been installed on it. This is due to lack of OpROM on XP941. The new Samsung SM951 been declared as "Supports Standard AHCI driver" though.