Very grateful that you posted this second look. Very interesting results. Now we just NEED a review of Fusion IO's Octal drive and my life will be complete. Make it happen Anand!
Might want to consider using the same scale on the graphs for each drive, clicking from P3700 to S3700 on the first graph, it was easy to miss the change in scale on the vertical axis. Would be easier to see difference if higher was faster across all drives.
In any event, I do really like this way of showing comparison data. keep up the good work!
Same. Especially true when the units are a one with a lot of zeroes but no commas, which makes it easy to mistake say 100,000 for 1,000,000 in the first place. I'd say go wild and go for both consistent scale AND commas!
sure no windows booting? maybe one simply have to have the right windows version (8.1 or server) and the right driver... see "NVM Express Boot Support added to Windows 8.1 and Windows Server 2012 R2" http://www.nvmexpress.org/blog/nvm-express-boot-su...
Quote from link: "Using the Windows inbox driver requires the platform to have the appropriate BIOS support and the device is not an eDrive implementation."
So BIOS upgrade should do it atleast? Here's hoping they'll fix BIOS´s
so there seem to be two differnt drivers available: one Windows included and one from Intel. Each seem to give a completly different performance for this Device: see here: http://www.pcper.com/reviews/Storage/Intel-SSD-DC-... (at least with the tested versions) => the question is: can we have both already TODAY? Great Performance AND Boot support (using the same driver)?
In the 25% OP 4KB Random Write (QD32) graph, the Samsung XP941 showed a massive drop between its 512GB and 384GB (25% spare area) results. From 30k down to 5k - is that an anomaly?
Also, what's with the vertical scale showing 1, 10, and 100 ranges? That forces all of the data points into a much smaller range, and it's visually not as informative.
I see - the two charts on the bottom of page one are same but with different vertical scales (as explained so well by DanNeely below).
Yes, I suppose instead of calling it a drop, you could say it's actually a rise from 5k to 30k IOPs when you go 25% spare area. It seems Samsung drives especially like extra spare area.
It's called a log scale; and generally is the better choice when you've got values spanning more than 2 or 3 orders of magnitude (ie more than 100x or 1000x). A linear scale would crush everything but the largest values into a tiny area at the bottom. With the highly consistent performance of the DC P3700 it's not as big a concern; but with the less consistent behavior of most consumer or older enterprise drives it's the only good way to see the behavior. Look at the Samsung 840; it has peak IOPS of ~100,000; drops to a steady state behavior mostly between ~3000-9000 but with occasional excursions as low as ~110.
Also, the tests are being done to show consistency of performance, a log scale makes seeing that much easier because the same vertical distance between min and max always represents the same multiple (eg 2x 3x 10x) between the values; a linear scale would mask inconsistency in slower drives while exaggerating it in higher performing ones because because 20000-22000 is a much larger interval on a linear scale than 10-1000 despite the former being a highly consistent drive with 10% variation and the latter having a 100x variation.
I'm enjoying the pivotchart type graphs you guys transitioned to a while ago. Makes readability and comparisons a breeze without the clutter, thanks for the change!
why don't you test the same drives also with 25% OP in storage bench 2013? samsung 840 pro (25% OP) vs sandisk extreme II (25% OP)...factory settings seem to penalize some drives.
I thought the same thing. The 2013 storage bench writes an insane amount of data in a short time. If someone really cares about performance with such a heavy write-load, then they would certainly overprovision their SSD. The P3700 obviously already comes from the factory with a lot of OP. It only makes sense to also OP the Samsung 840 Pro and other drives, as a point of comparison, when running the 2013 Storage bench.
To be clear, I mean to run the 2013 Storage Bench on each SSD with and without OP.
If I had money burning a hole in my pocket and there were computers available that could support these new SSDs I'd be buying one as soon as possible ... especially the P3700. Love those endurance numbers as much as the speed benchmarks. I can't wait until there is intense competition for these NVMe drives and the prices for 500 GB+ 3700s become as affordable as the Samsung 840 Pro SSD etc.
Looks like the days of being irritated having to twiddle your thumbs looking at an hourglass while some program or game loads is about to be over with in the next few years. Near instantaneous full reboots of Windows. Awesome just thinking about it. THIS is what will amaze people once everyone can afford these NVMe drives. I think these SSDs will be a huge hit. Just gotta get that freakin price down!
I can see it now. The year is 2020. Joe Schmo unpacks his brand new screaming fast Dell with a top notch PCIE SSD. He gets out the timer and boots it up. Total boot time is 47 seconds, with 42 of that being the BIOS load. Then it only takes 45 more seconds for the wireless card to connect to the network. Yes.
hopefully the PCIe is on PCIe 4 and the time is down to 4.5 sec total boot to desktop/network ready time. If Windows 9 is worse than Windows 8 - all bets are off!
Windows 8x - The Worst! of BOTH worlds! - for the price of 1/2 (or is it no charge already?).
Impressive. The earliest tests didn't look at that hot for client work, but these sure do, and match up with PCPer's results. Now we just need to see the <$0.75/GB consumer models, how they fare, and how quickly NVMe support matures. If it's anything like AHCI was for HDDs, 2016 will be a killer year for storage :).
We’ve updated our terms. By continuing to use the site and/or by logging into your account, you agree to the Site’s updated Terms of Use and Privacy Policy.
36 Comments
Back to Article
alacard - Tuesday, June 10, 2014 - link
Very grateful that you posted this second look. Very interesting results. Now we just NEED a review of Fusion IO's Octal drive and my life will be complete. Make it happen Anand!extide - Tuesday, June 10, 2014 - link
x2 on both! Love this drive though!lukewayne - Tuesday, June 10, 2014 - link
Anand,Great content as usual.
Might want to consider using the same scale on the graphs for each drive, clicking from P3700 to S3700 on the first graph, it was easy to miss the change in scale on the vertical axis. Would be easier to see difference if higher was faster across all drives.
In any event, I do really like this way of showing comparison data. keep up the good work!
nfriedly - Tuesday, June 10, 2014 - link
I was going to say the same thing - my initial impression was that it was about on par with the other drives.brucek2 - Tuesday, June 10, 2014 - link
Same. Especially true when the units are a one with a lot of zeroes but no commas, which makes it easy to mistake say 100,000 for 1,000,000 in the first place. I'd say go wild and go for both consistent scale AND commas!mtoma - Tuesday, June 10, 2014 - link
Is this drive bootable?althaz - Tuesday, June 10, 2014 - link
In OSX and Linux yes, but AFAIK not yet in Windows.hpvd - Tuesday, June 10, 2014 - link
sure no windows booting?maybe one simply have to have the right windows version (8.1 or server) and the right driver...
see
"NVM Express Boot Support added to Windows 8.1 and Windows Server 2012 R2"
http://www.nvmexpress.org/blog/nvm-express-boot-su...
juhatus - Tuesday, June 10, 2014 - link
Quote from link: "Using the Windows inbox driver requires the platform to have the appropriate BIOS support and the device is not an eDrive implementation."So BIOS upgrade should do it atleast? Here's hoping they'll fix BIOS´s
Sunburn74 - Tuesday, June 10, 2014 - link
Being bootable is a big deal and a must have featurehpvd - Tuesday, June 10, 2014 - link
if OSX and Linux Boot is already possible, the Bios should already be fine !?DanNeely - Tuesday, June 10, 2014 - link
Doesn't OSX only support UEFI?xMoe - Sunday, June 15, 2014 - link
OSX on Intel supports EFI - IIRC - though Apple may have a slightly different "flavor" ... 1058 and lower support Open Firmware - if you have an old rig -if you - or anyone is interested - here is more info ---
http://en.wikipedia.org/wiki/Apple%E2%80%93Intel_a...
http://en.wikipedia.org/wiki/EFI_System_partition
http://wiki.osx86project.org/wiki/index.php/Apple&... - last updated December 02, 2011.
http://en.wikipedia.org/wiki/Extensible_Firmware_I... <- strangly this lacks any specifics on Apple tech and/or standard (or lack there of).
http://en.wikipedia.org/wiki/GUID_Partition_Table <- slightly OT - but going over drive partitions - tables MBR, GPT, etc...
althaz - Tuesday, June 10, 2014 - link
Aha, I hadn't tried it since 8.1 came out, didn't realise they'd added it :). Awesome!hpvd - Tuesday, June 10, 2014 - link
so there seem to be two differnt drivers available: one Windows included and one from Intel.Each seem to give a completly different performance for this Device:
see here: http://www.pcper.com/reviews/Storage/Intel-SSD-DC-...
(at least with the tested versions)
=> the question is: can we have both already TODAY?
Great Performance AND Boot support (using the same driver)?
hpvd - Tuesday, June 10, 2014 - link
and which driver and which version is used by anandtech for reviewing P3700?Ryan Smith - Tuesday, June 10, 2014 - link
The Intel driver. The latest (and currently only) version: 1016sorten - Tuesday, June 10, 2014 - link
I don't see any reference to OSX here:http://en.wikipedia.org/wiki/NVM_Express#Operating...
UltraWide - Tuesday, June 10, 2014 - link
The barrier to entry is the high price. Everything else is top notch.nathanddrews - Tuesday, June 10, 2014 - link
Hence, the P3600 and P3500.[-Stash-] - Tuesday, June 10, 2014 - link
Will be great to see how the 35 and 36 perform in the client workloads – really quite excited about these.romrunning - Tuesday, June 10, 2014 - link
In the 25% OP 4KB Random Write (QD32) graph, the Samsung XP941 showed a massive drop between its 512GB and 384GB (25% spare area) results. From 30k down to 5k - is that an anomaly?Also, what's with the vertical scale showing 1, 10, and 100 ranges? That forces all of the data points into a much smaller range, and it's visually not as informative.
Kristian Vättö - Tuesday, June 10, 2014 - link
Are you sure you are not mixing up the graphs? With 25% spare area, the results are better (~30K IOPS), which is how things should be.romrunning - Tuesday, June 10, 2014 - link
I see - the two charts on the bottom of page one are same but with different vertical scales (as explained so well by DanNeely below).Yes, I suppose instead of calling it a drop, you could say it's actually a rise from 5k to 30k IOPs when you go 25% spare area. It seems Samsung drives especially like extra spare area.
DanNeely - Tuesday, June 10, 2014 - link
It's called a log scale; and generally is the better choice when you've got values spanning more than 2 or 3 orders of magnitude (ie more than 100x or 1000x). A linear scale would crush everything but the largest values into a tiny area at the bottom. With the highly consistent performance of the DC P3700 it's not as big a concern; but with the less consistent behavior of most consumer or older enterprise drives it's the only good way to see the behavior. Look at the Samsung 840; it has peak IOPS of ~100,000; drops to a steady state behavior mostly between ~3000-9000 but with occasional excursions as low as ~110.Also, the tests are being done to show consistency of performance, a log scale makes seeing that much easier because the same vertical distance between min and max always represents the same multiple (eg 2x 3x 10x) between the values; a linear scale would mask inconsistency in slower drives while exaggerating it in higher performing ones because because 20000-22000 is a much larger interval on a linear scale than 10-1000 despite the former being a highly consistent drive with 10% variation and the latter having a 100x variation.
ryanjacoby2 - Tuesday, June 10, 2014 - link
I'm enjoying the pivotchart type graphs you guys transitioned to a while ago. Makes readability and comparisons a breeze without the clutter, thanks for the change!uruturu - Tuesday, June 10, 2014 - link
why don't you test the same drives also with 25% OP in storage bench 2013?samsung 840 pro (25% OP) vs sandisk extreme II (25% OP)...factory settings seem to penalize some drives.
morganf - Tuesday, June 10, 2014 - link
I thought the same thing. The 2013 storage bench writes an insane amount of data in a short time. If someone really cares about performance with such a heavy write-load, then they would certainly overprovision their SSD. The P3700 obviously already comes from the factory with a lot of OP. It only makes sense to also OP the Samsung 840 Pro and other drives, as a point of comparison, when running the 2013 Storage bench.To be clear, I mean to run the 2013 Storage Bench on each SSD with and without OP.
eanazag - Tuesday, June 10, 2014 - link
When's the Intel giveaway for the P3700?extide - Tuesday, June 10, 2014 - link
LOL! Yeah, SIGN ME UP!! Haha ;)TelstarTOS - Tuesday, June 10, 2014 - link
+1TelstarTOS - Tuesday, June 10, 2014 - link
YES! Thank you for doing these further tests :)aperson2437 - Tuesday, June 10, 2014 - link
If I had money burning a hole in my pocket and there were computers available that could support these new SSDs I'd be buying one as soon as possible ... especially the P3700. Love those endurance numbers as much as the speed benchmarks. I can't wait until there is intense competition for these NVMe drives and the prices for 500 GB+ 3700s become as affordable as the Samsung 840 Pro SSD etc.Looks like the days of being irritated having to twiddle your thumbs looking at an hourglass while some program or game loads is about to be over with in the next few years. Near instantaneous full reboots of Windows. Awesome just thinking about it. THIS is what will amaze people once everyone can afford these NVMe drives. I think these SSDs will be a huge hit. Just gotta get that freakin price down!
seapeople - Tuesday, June 10, 2014 - link
I can see it now. The year is 2020. Joe Schmo unpacks his brand new screaming fast Dell with a top notch PCIE SSD. He gets out the timer and boots it up. Total boot time is 47 seconds, with 42 of that being the BIOS load. Then it only takes 45 more seconds for the wireless card to connect to the network. Yes.xMoe - Sunday, June 15, 2014 - link
hopefully the PCIe is on PCIe 4 and the time is down to 4.5 sec total boot to desktop/network ready time. If Windows 9 is worse than Windows 8 - all bets are off!Windows 8x - The Worst! of BOTH worlds! - for the price of 1/2 (or is it no charge already?).
Cerb - Saturday, June 14, 2014 - link
Impressive. The earliest tests didn't look at that hot for client work, but these sure do, and match up with PCPer's results. Now we just need to see the <$0.75/GB consumer models, how they fare, and how quickly NVMe support matures. If it's anything like AHCI was for HDDs, 2016 will be a killer year for storage :).